Lunar craters with cracked floors

Item Type text; Thesis-Reproduction (electronic)

Authors Weller, Roger Nelson, 1944-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 24/09/2021 00:53:28

Link to Item http://hdl.handle.net/10150/347802 LUNAR CRATERS WITH CRACKED FLOORS

by

Roger Nelson Weller

A Thesis Submitted to the Faculty of the

DEPARTMENT OF GEOSCIENCES

In Partial Fulfillment of the Requirements For the Degree of

MASTER OF SCIENCE

In the Graduate, College

THE UNIVERSITY OF ARIZONA

19 7 2 STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of re­ quirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judg­ ment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

SPENCER R. TIT LEY Date Professor of Geosciences ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my many friends and colleagues who have donated much valuable time in assisting me with this thesis. I am especially indebted to Dr. S. R. Titley for criti­ cally reviewing the thesis, to Dr. W. B. Bull for stressing the develop­ ment of interpretative criteria, and to Dr. R. G. Strom for evaluating the feasibility of the proposed mechanisms. Other workers from whose constructive criticisms and us.eful suggestions I have greatly benefited are E. A. Whitaker and Drs. P. E. Damon, W. K. Hartmann, G. P.

Kuiper, W.C. Lacy, and E. B. Mayo.

I would also like to thank my fellow students, J. Cannon, N.

Colburn, R. Holcomb, P. Kresan, S. Larson, D. Vukobratovich, and

H. Welhener not only for allowing me to sound out my ideas but also for their assitance in typing and in the preparation of photographs. I am especially indebted to my parents for their kind encourage­ ment in my studies throughout my youth and for their faith in me. I would also like to acknowledge my thankfulness to two people who were influ­ ential in directing my interest at any early age toward geology: Mr.

Walter Bone and my grandmother, Mrs. Arthur Weller.

A portion of this study was financed by a summer trainee ship in 1970 from the National Science Foundation. TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS...... vil LIST OF TABLES ...... xii

ABSTRACT...... xiii

1. INTRODUCTION...... 1

The Problem ...... 1 The Approach . 1 Materials and Conventions ...... 4 2 . ' DESCRIPTION AND GENERAL NATURE OF THE CRACKS . . . . 5 Crack W idths ...... 5 Maximum and Mean Maximum Crack Widths .... 5 Mean Crack W idth ...... 8 Minimum Crack Width . 8 Crack Depth and Wall S l o p e s ...... 9 Summary of Crack Cross-sectional P r o f ile s...... 12 Total Crack Lengths and Crack Density ...... 13 Crack Direction ...... 14 Rose Diagrams. . 16 Crater-centered Directional C ontrol...... 20 Crack Patterns...... 21 Pitatus Crack Pattern...... 21 Humboldt Crack Pattern...... 29 Petavius Crack Pattern...... 29 Alphonsus Crack Pattern ...... 29 Lavoisier H Crack Pattern . 29 Wurzelbauer Crack Pattern ...... 38 Summary of the Crack Patterns ...... 38 Crack Configurations...... 38 Summary of Crack Descriptions ...... 40

3. CLASSIFICATION AND GENESIS OF THE LUNAR RILLES . . . . 46

Sinuous Rilles...... 46 Wide-floored Rilles . 52 Irregular Rilles ...... 60 Lineament Rilles...... 64 Cracks ...... 66

iv V

TABLE OF CONTENTS— Continued

• . Page 4. CRATER-FLOOR MATERIALS ...... 69 Original Crater-floor Materials ...... 69 Dark Secondary Crater-floor Fillings 72 Light-colored Secondary Crater Fillings ...... 79 Small-scale Features of Possible Volcanic Origin. . . . 85 - Summary of Crater-floor Materials ...... 92

5. CRATER FLOOR AND RIM MORPHOLOGY , . . . , ...... 93 Crater-floor Morphology ...... 93 Crater-rim Morphology ...... 101 6. REGIONAL CONTROL OF CRATERS WITH CRACKED FLOORS . . 109

Size-frequency Distribution of Lunar Craters with Cracked Floors ...... 109 Distribution of Craters with Cracked Floors by Latitude and Longitude...... 112 Distribution by Latitude .' ...... 112 Distribution by Longitude ...... 113 Distribution of the Craters with Cracked Floors in Relation to the Maria...... 117 Formation of the Circular Maria ...... 119 Formation of the Irregular Maria ...... 126 Lineaments Related to Craters with Cracked Floors . . . 127 Summary ...... 130

7. TERRESTRIAL ANALOGUES ...... 133 Resurgent Calderas...... 134 Terminology and General Characteristics ...... 134 Valles Caldera...... 136 Timber Mountain Caldera ...... 138 Turkey Creek Caldera...... 142 Comparison between Terrestrial Resurgent Calderas and Lunar Craters with Cracked Floors . 143 Cauldron Block vs Crater Filling ...... 145 Physical Characteristics of Magmas. Associated with Lunar Craters with Cracked Floors ...... 146 Martian Analogues to the Lunar Craters with Cracked Floors ...... 148 vi

TABLE OF CONTENTS--Continued

Page 8„ A GENERALIZED MODEL FOR THE FORMATION OF LUNAR CRATERS WITH CRACKED FLOORS ...... 150

Crater Formation ...... 150 Crater Fillings...... 154 Development of the Cracks ^ ...... 157 Implications...... 159 APPENDIX I: LUNAR CRATERS WITH CRACKED FLOORS . . . 161 APPENDIX II: WIDTH MEASUREMENTS OF CRACKS . . . . . 165

APPENDIX III: DEGREE OF CRACKING...... 168

APPENDIX IV: CRACK PATTERNS ...... 171

REFERENCES...... 173 LIST OF ILLUSTRATIONS

Figure Page 1.' The Lavoisier Group of Craters with Cracked Floors . . . 2 2. Crater Palmier! ...... • • • • . V ...... 3 3. Maximum and Mean Maximum Crack Width Values Plotted vs Crater Diameter ...... 6

4. Mean and Minimum Crack Width Values Plotted vs Crater Diameter ...... „ ... 7

5. Hypothetical Models of Shadow Geometry in Cracks . . . 10

6. Close-up View of a Typical Crack within the Crater Gassendi ...... 11 7. Crater Density Plotted vs Crater Diameter ...... 15

8. Crater Hevelius ...... 17

9. Map of Hevelius Derived from Figure 8...... 18

10. Crater Lavoisier D ...... 19

11. Semi-quantitized Crack Densities for Five Angular Crack Relationships ...... 22

12. Graphic Summary of Idealized Major Crack Patterns . . . 23

13. Craters Pitatus, Hesiodus , and Wurzelbauer ...... 24

14. Map of Pitatus Region Derived from Figure 13 ...... 25

15. Craters Lavoisier and Lavoisier F ...... 26 16. The Lunar Far side Crater Chappell ...... 27 17. The Lunar Farside Basin-crater Oppenheimer ...... 28

18. Crater Humboldt...... 30

19. Map of Humboldt Derived from Figure 18 ...... 31

20. Crater Schluter ...... 32

v ii . v iii

LIST OF ILLUSTRATIONS—Continued

Figure Page

21. Grater Petavius ...... 33 22. Craters Einstein "V" and Einstein " U" ...... 34 23. Crater Alphonsus 35

24. Map of Alphonsus Derived from Figure 23 ...... 36

25. Crater Lavoisier H, ,, .... „ ...... 37

26. Eastern Portion of the Crater Floor in Humboldt ..... 39 27. Northeast Portion of the Crater Floor in Pitatus ..... 41

28. Crater Arzachel ...... 42 29. Craters Rep sold and Repsold G ...... 43

30. Crater Messala ...... 44

31. Crater Posidonius ...... 48

32. Map of Posidonius Derived from Figure 31 ...... 49

33. Schroter's Valley and the Cobra Head ...... 50

34. Wide-floored Rilles and Strings of Secondary Impact Craters near the Crater Ramsden ...... 54 35. Wide-floored Rilles Adjacent to the Crater Campanus . . 55

36. Trace Configurations Characteristic of Wide-floored Rilles ...... 56.

37 . . Eastern Margin of Mare Serenitatis ...... 58

38. Crater Hyginus and the Hygihus Rille ...... 59 39. Trace Configurations Characteristic of Irregular Rilles . . 62

40. Wide-floored and Irregular Rilles near the Crater Triesnecker ...... 63

41. Lineament Rilles near the Crater Airy...... 65 42. Rough Floor Materials in the Central Portion of Tycho . . . 71 ix

LIST OF ILLU STRAT IONS --Continued

Figure Page 43. A Close-up View of the Northern Portion of the Crater Floor in TsioIkovsky ...... 74

44. Modified Cracks and Associated Dark Floor Materials Within the Crater Petavius . ^ . .- . . . . 76 45. Dark-halo Craters on the Western Portion of the Floor of Alphonsus...... 77

46. Thick-rimmed, Light-colored Craters along Cracks in Alphonsus ...... 81 47. The Central Section of the Floor of Gassendi ...... 83

48. Hummocky Floor Materials Within the Crater Petavius . . 84

49. Eastern Portion of the Floor of Hesiodus . 87

.50. Map of Eastern Portion of Hesiodus Derived from . Figure 49 ...... 88

51. Crater Mersenius ...... 90

52. Map of Mersenius Derived from Figure 51 ...... 91

53. Updoming and Partial Collapse of the Brittle Floor Plate . . 96 54. Crater Hansteen ...... 97

55. Crater La Condamine...... 98

56. Uplift of a Thick Brittle Plate and . Subsidence of a Brittle Plate...... 99

57. Crater " Mare Smythii D" ...... 100

58. Crater Cardanus...... 102

59. Crater Briggs ...... 104

60. Crater Kopf...... 105

61. Crater Schluter "X" 106 62. Crater Damoiseau ...... 107 X

LIST OF. ILLUSTRATIONS—Continued

Figure Page 63. Lunar Frontside Size-frequency Distribution of Craters with Cracked Floors ...... 110 64. Calculations for Standardizing Areas, within. 10° Latitude Belts ...... 114 65. Corrected Latitude Distribution of Lunar Frontside Craters with Cracked Floors ...... 115

66. Longitude Distribution of Lunar Frontside Craters with Cracked Floors ...... 116

67. Map of the Frontside of the Moon Showing the Distribution of Craters with Cracked Floors . . .in pocket

68. Mare Moscoviense ...... 118 69. Northwestern Margin of Oceanus Procellarum ...... 120 70. Map of Northwestern Margin of Oceanus Procellarum Derived from Figure 6 9 ...... 121

71. Mare Smythii ...... 122

72. Mare Orientale...... 124 73. The Lavoisier Group of Lunar C raters...... in pocket

74. Crater Einstein A and Eight Associated Minor Craters with Cracked Floors ...... 128

75. Map of Crater Einstein A and Eight Associated Minor Craters with Cracked Floors Derived from Figure 74 . . 129.

76. . Craters Einstein "W" and Einstein "X" ...... 131

77. Geologic Map of the Valles Caldera, New Mexico . . . . 137

78. Geologic Map of the Timber Mountain.Caldera...... 140

79. An Oblique View of Damoiseau ...... 147

80 . Idealized Structure of Large Impact Crater ...... 153

81. Development of a Chilled Crust over a Post-impact Lava Crater Filling ...... 156 xi

LIST OF ILLUSTRATIONS— C ontinued

Figure Page.

82. Updoming of the Crater Floor by Laccolithic Intrusion and the Emplacement of Moat Volcanics . . 158

( / LIST OF TABLES

Table Page

1. Characteristics of Four Major Lunar Rille Types ..... 67

2. Examples of Dark Crater-floor Materials Found in Craters Which Contain Cracks ...... 79

3. Examples of Various Crater-floor Profiles of Craters Containing Cracks ...... 94

4. Examples of Craters with Cracked Floors That Are Associated with the Major Lunar Maria ...... 117 5. Examples of Terrestrial Resurgent Cauldrons and Caldcras...... 136

6. Structural and Volcanic Development of the Valles Caldera, New Mexico ...... 139

7. Summary of Features Common to Terrestrial Resurgent Calderas and Lunar Craters with Cracked Floors ...... 144

x ii ABSTRACT

Concentrations of large craters whose floors contain cracks occur within several restricted regions of the lunar surface. Typically, these cracks are restricted to the crater floors and have low sloping walls. Intersecting, terminating, and branching configurations, as well as radial and concentric patterns, are displayed by these cracks, sug­ gesting tensional deformation of brittle materials on the crater floors.

A review of the major types of.lunar rilles also indicates a tensional origin for the cracks. Crater-floor materials associated with the cracks are quite diverse in tone and morphology , implying a complex mode of emplacement for these materials, while the crater rims suggest both meteoritic impact and volcanic origins for the initial depressions. Many lunar craters with cracked floors occur along the margins

Of the irregular and circular maria and appear to be associated with the tensional lineaments surrounding these maria. It is proposed that the craters with cracked floors were originally flooded by volcanic materials erupting from circum-mare fissures early in the formation of the maria and that these crater fillings were later cracked by subsequent volcanic activity along the same fissures.

Most lunar craters with cracked floors are comparable both in size and morphology to silicic terrestrial resurgent calderas, suggesting that the occurrence of structures similar to the lunar craters with cracked floors might be anticipated on any planet (for example. Mars) where a ' . - xiv large depression is found superimposed upon a major regional- lineament along which prolonged, viscous volcanic activity has occurred. CHAPTER I

INTRODUCTION

The Problem '

Within certain regions of the lunar surface there are concentra- tions of large craters, often over 30 kilometers in diameter, whose floors are crossed by complex patterns of intersecting rilles (Figure 1) „ This thesis is an investigation into the genesis.of the rilles on crater floors „ By studying morphologic features and stratigraphic relationships, one can deduce possible mechanisms for rille production and apply these hypotheses to lunar mare formation.

A brief examination of the rilles within craters reveals that they are generally restricted to the crater floors; such rilles will be referred to as "cracks" throughout this thesis. The rilles found within Palmier!

. (Fig. 2) do not qualify as cracks because they extend beyond the crest of the crater rim.

v The Approach

A reasonable approach to many observational studies is to begin by observing and describing the basic elements of the problem. Within craters with cracked floors, the basic structural elements are the cracks; these features are described in chapter 2. The next step is a considera­ tion of the interrelationships of the basic elements; in chapter 3, after a comparison between the cracks and other lunar rilles, a tentative model of crack formation is formulated. Attention is then directed in chapter 4 Figure 1. The Lavoisier Group of Craters with Cracked Floors 3

Figure 2. Crater Palmier! to the materials in which the cracks occur„ Cross-sectional profiles of crater floors are evaluated in chapter 5 as possible indicators of vertical movement. Next, regional control of crater distribution is considered in chapter 6. To test the validity of the hypotheses, terrestrial structures analogous to the lunar craters with cracked floors are reviewed in chap­ ter 7. Finally, hypothetical models for cracking are summarized and important consequences suggested.

Materials and Conventions

Most of the photographs employed throughout this thesis were derived from the Lunar Orb iter missions IV and V. With the exception of albedo photographs from the Rectified Lunar Atlas (Whitaker et a l., 1964), terrestrial photography was of little use due to its limited resolution.

Ranger photography provided close-up views of the eastern portion of the crater floor in Alphonsus, and Apollo photography was restricted to a narrow equatorial belt in which a limited number of craters with cracked floors occur. Coordinates of cracked craters were taken from the tables pre­ sented in "The System of Lunar Craters," quadrant I (Arthur et a l . , 1963), quadrant II (Arthur et a l., 1964), quadrant III (Arthur et a l., 1965), and quadrant IV (Arthur, Pellicori, and Wood, 1966), published in the

Communications of the Lunar and Planetary Laboratory. Crater diameters were calculated from measurements made by the author of Lunar Orb iter IV high-resolution photographs. CHAPTER 2

DESCRIPTION AND GENERAL NATURE OF THE CRACKS

The geometry of the cracks can be described by two sets of physical parameters „ One set is derived from the cross-section view of a crack and consists of measurements of width, depth, and wall slope.

The other set describes the variable nature of a crack along its trace in terms of length, direction, position, and characteristic configurations.

Crack Widths Four different values of crack width can be obtained for a par­ ticular crater: the maximum, the mean maximum, the mean, and the minimum crack width. Measurements were made of these values for 47 lunar frontside craters which were chosen on a basis of clarity from the high-re solution Lunar Orb iter IV photography. Results for the individual craters are listed in Appendix II, and the data are graphically plotted versus crater diameter in Figures 3 and 4.

Maximum and Mean Maximum Crack Widths The maximum and mean maximum crack widths are both obtained from the widest crack within a particular crater. The maximum crack width is the largest separation found along the widest crack, whereas the mean maximum crack width is obtained by averaging five width measurements from points equally spaced along the crack. The maximum crack width establishes a definite upper limit on how wide a crack might

5 CRACK WIDTH (KM)

4.0

3 . 0 -

2 .0 - -

0.0 0 20 40 60 80 iOO 120 140 160 ISO 200 CRATER DIAMETER (KM) a. Maximum Crack Width vs Crater Diameter

CRACK WIDTH (KM)

4.0 --

3.0 --

2 .0 - -

1 . 0 --

0.0 20 40 60 80 100 120 140 160 ISO 200 CRATER DIAMETER (KM) b. Mean Maximum Crack Width vs Crater Diameter

Figure 3. Maximum and Mean Maximum Crack Width Value Plotted vs Crater Diameter CRACK WIDTH (KM)

4.0 --

3.0 -

2.0 - €£>

1.0 -• © d9 qgo ®

0.0 0 20 40 60 80 100 120 140 160 180 200 CRATER DIAMETER (KM) a. Mean Crack Width vs Crater Diameter

CRACK WIDTH (KM)

4.0 --

3.0

2.0 -•

1.0

0.0 20 40 60 80 100 120 140 160 180 200 CRATER DIAMETER (KM) b. Minimum Crack Width vs Crater Diameter

Figure 4. Mean and Minimum Crack Width Values Plotted vs Crater Diameter be, while the mean maximum crack width describes the average width.of the widest crack „ In reference to Figures 3a and 3b, both curves are observed to increase slowly with increasing crater diameter. This relationship may reflect the thickness of crater-floor materials or it may represent the ex­ ponential nature of the probability function that relates crack-width frequency to area. The data also suggest an effective upper limit for crack widths at 2.5 km.

Mean Crack Width

The mean crack-width value was obtained by taking the average of ten crack-width measurements of separate cracks found within a par­ ticular crater, or, if only a small number of cracks were present, mea­ surements were taken instead at points equally spaced along the available cracks. These data, shown in Figure 4a, indicate a negligible increase in crack width with increasing crater diameter. These results imply that the mean crack width can be considered to be effectively in­ dependent of crater diameter or floor area. In Appendix 11, the overall mean crack width for 47 craters is 0.72 km.

Minimum Crack Width

The minimum crack width is the smallest measurement of crack width that can be obtained for a particular crater. Figure 4b shows this value to be independent of crater diameter, and from Appendix II a mini­ mum crack width of 0.4 km was derived. Inspection of high-resolution Lunar Orb iter V photographs of Petavius and Alphonsus show that smaller cracks can exist. Therefore, 0.4 km is probably the effective 9 detectability of cracks on the high-re solution Lunar Orb iter IV photog­ raphy.

Crack Depth and Wall Slopes From a geometrical analysis of shadows within cracks, two more physical parameters can be determined: the crack depth and wall slopes. A 1.3-km-wide crack in Posidonius was chosen as a test case because of the 23°' sun angle present in frame 086-3H of the Lunar Or- biter IV photography. This crack trends approximately north-south

(perpendicular to the sun's rays) and appears to be a rather typical crack. The shadow from the eastern crack wall falls approximately in the center of this crack. By assuming bilateral symmetry for the crack walls and considering the various geometric profiles shown in Figure 5, one is forced to choose a profile similar to example 5c. Given the sun angle (23°) and the crack width , a crack depth of 250 meters is derived; in other words, the depth of this crack is only one-fifth of its width.

Since similar shadow geometries are found in most cracks observed in

Lunar Orb iter IV photographs, a depth-to-width ratio of 1 to 5 may be used as a general model for most cracks. The cross-sectional profile most commonly encountered in cracks is a modified "U" shape, as shown in Figure 6, Steeps walls and flat floors are rarely observed. Keeping within these boundaries, one can devise a sequence of possible cross-sectional profiles, ranging from a uniform slope in which the steeper slope angles occur just below the crack rim. Reconsidering shadow geometries and Figure 5c, one 10

BASIC GEOMETRY

HIGHLIGHT SHADOW

DEEP FISSURE GEOMETRY

HIGHLIGHT SHADOW

NORMAL CRACK HIGHLIGHT SHADOW GEOMETRY

Figure 5. Hypothetical Models of Shadow Geometry in Cracks 11

Figure 6. Close-up View of a Typical Crack within the Crater Gassendi 12 would need to limit the uniform slope to 23°, but might also assume short slopes as steep as 45° for the second model.

The range of slope angles predicted for crack walls encom­ passes the angle of repose for fine, unconsolidated materials in the lunar environment. Within the close-up view of the crack shown in

Figure 6, no rock outcrops are observed. Instead, the crack walls are patterned with the "tree-bark" textured surface which has often been attributed to down slope creep in fine lunar materials. The lack of small sharp-rimmed craters on the crack floors and walls also suggests a fine, unconsolidated cover. High-re solution Lunar Orbiter V views of cracks within Petavius, Alphonsus, and Doppelmayer revealed similar textures and profiles. Lowman (1969) suggests that the high albedo of the crack walls seen within Humboldt are due to mass wasting which exposes fresh materials. It appears then that the shallow cross-sectional pro­ files of the cracks are at least due in part to mass wasting of fine, unconsolidated materials.

Summary of Crack Cross-sectional Profiles

In general, the cracks within craters are variable in width, ranging from a maximum of 2.5 km to values just below the effective detectability limit (o.4 km) of the high-resolution Lunar Orbiter IV pho­ tography. The mean crack width obtained for 47 craters was 0.72 km.

From the geometrical analysis of shadows within cracks, a mean slope of 23° is derived for a typical crack, resulting in a crack- depth-to-width ratio of 1 to 5. From this ratio, one would expect a 1- km-wide crack to be approximately 200 meters deep. 13 The crack-wall slopes, the "tree-bark" textured surface found on the rims and walls of the cracks , the absence of exposed rock out­ crops, and the lack of small sharp-rimmed craters all tend to suggest that the present morphology of the cracks is partially due to mass wast­ age of relatively fine, unconsolidated materials.

Total Crack Lengths and Crack Density

The first property needed to describe crack traces is a measure­ ment of crack length. Since the cracks are multiple, interrelated features, individual crack lengths often cannot be calculated. Instead, a sum of the crack lengths within a crater can be measured and from this value a related parameter, the crack density, can be derived. Determination of the total length of cracks within a particular lunar crater is hindered by several difficulties. One problem is crack orientation. Linear depressions that trend nearly east-west are parallel to the sun's rays and therefore do not cast shadows even under condi­ tions of low-angle lighting. Another problem is the limited coverage of the Lunar Orb iter IV photography in which only one lighting angle was employed. Consequently, many cracks could be obscured by shadows along the eastern rims of the craters. Although actual distinctions may be somewhat arbitrary, features like crater chains and narrow terraces bounded by depressions should not be considered as cracks, although these features often merge with cracks. The author recognizes this dif­ ficulty and has tried to remain consistent with his designations.

Keeping in mind these difficulties, total crack-length measure­ ments were made for 48 craters. Results for the individual craters appear 14 in Appendix 3. The crater Balboa "X" (32.7 km in diameter) contained the shortest total length of cracks within the group of 48 craters, 11.8 km, while Humboldt (201.8 km in diameter), at the other extreme, con­ tained 1,373 km of cracks. The total crack length for a particular crater discloses little about the nature of the cracks. However, if the total crack length is expressed as a function of floor area, then relative densities of crack­ ing may be calculated. To simplify calculations, the author has made the assumption that crater-floor areas were directly proportional to the total area included within the rim of a crater; crack densities were then calculated using this larger area. Since only relative differences are being sought for,this substitution should affect all examples equally.

Significant discrepancies should arise only in rare instances where few floor materials exist or where craters appear to be filled to overflowing.

When the values of crack density are plotted versus crater diameter (Fig. 7), a complex relationship resu lts. The only generaliza­ tion that may be formed is that small craters have a higher probability of being more intensely cracked.

Crack Direction

The directional aspect of the cracks may be handled in two ways. In the more conventional approach, azimuthal directions are as­ signed to linear features and these directions are plotted on rose dia­ grams. The second, and less conventional, approach is to establish a radial coordinate system for each crater; this approach describes the angular relationships between the cracks and radial reference lines. CRACK DENSITY (KM/KM2 )

0.20 -- © O

0.15 --

0.10 --

0.05 -

0.00 0 20 40 6080 100 120 140 160 180 200 CRATER DIAMETER (KM) Figure 7. Crater Density Plotted vs Crater Diameter 16 Both approaches have slightly different goals. Rose diagrams are an at­ tempt to locate regional structural controls and thus favor the establish­ ment of tectonic grids, whereas the second approach isolates the indi­ vidual crater and tries to relate linear features to the crater's internal geometry and structure.

Rose Diagrams The usage of rose diagrams in recording crack directions en­ counters several inherent difficulties. Strictly speaking, the cracks are not linear features but are often arcuate or angular in their traces. To describe a particular crack, it would have to be approximated by arbi­ trarily determined straight-line segments. Lengths of segments are typically not recorded, just the number of linear features, so it is con­ ceivable that a long crack with several short bends would produce an unrelatable rose diagram. A rose diagram plot of lineaments obtained from a single photograph might also end being simply a plot of the func­ tion of crack detectability versus sun angle. As mentioned earlier, lunar rilles trending east-west are difficult to detect, while those trending north-south would be enhanced by the lighting.

Instead of using rose diagrams, the question of preferred region­ al directions of cracking can be approached by directly observing a few examples . Hevelius (Figs. 8 and 9) and Lavoisier D (Fig. 10) are the two craters that show the greatest degree of regional structural control of crack directions. Hevelius appears to be the better example of the two.

The three prominent crack directions found within this crater correspond quite closely to the three directions of the lunar tectonic grid, whose 17

m

I

%

5

Figure 8. Crater Hevelius 18

cr If m upu

cr

cc w cp

Hevelius If m

If m

hmk cch

upu ccw cch sc ccw

sc

Figure 9. Map of Hevelius Derived from Figure 8 Figure 10. Crater Lavoisier D 20 age predates the craters (Spurr, 1944; Fielder, 1965; Firsoff, 1961; and Strom, 1964). Lavoisier D also shows a strong system of cracks with weaker northeast and north-northeast systems.

However, the great majority of cracked Craters do not show regional trends. Instead, crack directions appear to be more closely related to the internal geometry of the craters. Pronounced radial and concentric crack patterns are the rule and not the exception, as verified by the Lavoisier group of craters (Fig. 1).

In summary, only a small portion of the cracks within craters are aligned with the older tectonic grid. The vast majority of cracks may be categorized as radial or concentric with respect to a crater.

Crater-centered Directional Control

Because the cracks in most instances appear to be related to the geometry of the crater, a technique had to be devised to evaluate this relation. Basically, it was assumed that every crater had a center of symmetry and that radial reference lines could be drawn from these centers, intersecting or paralleling cracks, depending upon the orienta­ tion of the cracks. Five angular relationships between radial reference lines and cracks were considered; (1) concentric cracks, which are near­ ly perpendicular to the radial reference lines; (2) anastomosing concen­ tric cracks, which are oriented slightly less than 90° and therefore often intersect each other at low angles; (3) intermediate-angle cracks, whose intersections create a crude, polygonal pattern; (4) subradial cracks, which are nearly parallel to the radial reference lines; and (5) radial cracks. 21 To assess the relative density of cracks within each angular category for a particular crater, a semi-quantitized graphic scale (Fig.

11) was created. The density of cracks falling into each angular rela­ tionship was then estimated from the scale and assigned a relative value decreasing in intensity from 4 to 0. The results obtained for 44 craters evaluated by this technique appear in Appendix IV. Mean crack densities were then calculated from these results for each of the five angular cate­ gories. Intermedlate-angle cracks turned out to be relatively rare. When the mean values are combined for the two concentric and two radial categories, a good-to-fair rating of 2.55 for the concentric element was obtained along with a weak-to-fair rating of 1.57 for the radial crack element. In summary, an average crater would be expected to have a denser concentration of concentric cracks in relation to its radial cracks.

Crack Patterns

Based on relative densities of radial and concentric elements and the zones of greatest crack density, six major crack patterns are observed on the floors of lunar craters. An idealized graphic summary of these patterns is shown in Figure 12.

Pitatus Crack Pattern

Lunar craters with a well-developed, peripherally located con­ centric crack component occur frequently. The crater Pitatus (Figs. 13 and 14) is the type crater for this particular pattern. Other prominent craters in this class are Lavoisier (Fig. 15), Chappell (Fig. 16), and

Oppenheimer (Fig. 17). In both Chappell and Oppenheimer, minor craters contain their own internally related cracks, a configuration that will be 22

STRONG (4) GOOD (3) FAIR (2) WEAK (I)

CONCENTRIC

ANASTOMOSING CONCENTRIC

INTERMEDIATE -ANGLE

SUB-RADIAL

RADIAL

Figure 11 . Semi-quantitized Crack Densities for Five Angular Crack Relationships HUMBOLDT

PETAVIUS ALPHONSUS

LAVOISIER H WURZELBAUER

Figure 12. Graphic Summary of Idealized Major Crack Patterns 24

Figure 13. Craters Pitatus , Hesiodus , and Wurzelbauer 25

- / /z ^ \' / l \

Hesiodus

' s d f m 1

X dfm . N Z / c c w

• \ \ V ; / ■rh\ ' - - y

/mrry ' V ^ i^ ) upu X 'I 1 \ v v'/' * s dfm K -/ / / x x vtv ^ Pitotus .

cc w

h rmk

cch/4 Z 'i i

^ y / y h v n k l , / cw i /

Wurzel bouer

Figure 14. Map of Pitatus Region Derived from Figure 13 Figure 15. Craters Lavoisier and Lavoisier F 27

Figure 16. The Lunar Farside Crater Chappell 28

Figure 17. The Lunar Farside Basin-crater Oppenheimer 29 discussed later in chapter 6 in relation to the development of the circu­ lar lunar maria „

Humboldt Crack Pattern

The craters Humboldt (Figs. 18 and 19) and Schluter (Fig. 20) typify another major crack pattern. In this pattern both the concentric and radial crack components are well developed . Unlike the Pitatus pat­ tern, concentric cracks within this group of craters are not confined to the margins of the crater floor but occur over the entire floor area.

Petavius Crack Pattern

The third distinctive crack pattern strongly emphasizes the radial crack components. Petavius (Fig. 21) is split into thirds by its radial cracks, while the floor of Einstein "V" (Fig. 22) is .divided into quarters.

Alphonsus Crack Pattern The crack pattern that appears within Alphonsus (Figs. 23 and 24) consists predominantly of subradial and anastomosing concentric cracks. The floors of Cleomedes and Darrioiseau also display this crack pattern.

Lavoisier H Crack Pattern

Lavoisier H (Fig. 25) contains an abundant number of interme­ diate angle Cracks which create a crude polygonal pattern. This pattern is commonly found among the smallest craters with cracked floors, such as Schluter "X" , and the small craters surrounding Einstein A and Eddington. Figure 18. Crater Humboldt 31

upu

cc w upu

df m

ot m

df m

If m ^ cch cc w cp

dfm ,v cc w

If m upu H. x. Humboldt

Figure 19. Map of Humboldt Derived from Figure 18 32

Figure 20. Crater Schluter 33

Figure 21. Crater Petavius Figure 22. Craters Einstein "V" and Einstein "U" 35

Figure 23. Crater Alphonsus 36

cc w Alphonsus

\v \ ccw I X

cOdm7

ccw

ccw \

/ upu

Figure 24. Map of Alphonsus Derived from Figure 23 37

Figure 25. Crater Lavoisier H 38 Wurzelbauer Crack Pattern

Many craters contain only a few cracks and therefore are not

readily assignable to any particular crack pattern. Wurzelbauer (Figs. 13 and 14) is a typical example of this phenomenon. Other craters con­ taining a limited number of cracks are Campanus, Lavoisier C, and

M essala.

Summary of the Crack Patterns

The crack patterns observed on the floors of lunar craters are highly variable „ No specific isolated patterns exist; examples of transi­

tional stages are found between all six of the highly idealized patterns just described. It may be stated in the defense of this classification, however, that the major proportion of lunar craters with cracked floors

resemble one of these six crack patterns „

Crack Configurations

The term "crack" appropriately describes the rilles found on

crater floors, for these rilles display trace configurations which are

characteristic of small-scale cracks in brittle materials. As an example,

the fracture pattern in Humboldt (Figs. 18 and 26) shows;

1. The termination of one crack by another ,

2. Pronounced radial and concentric patterns (as commonly ob­

served in shattered auto safety glass), and ,3 . The lack of strike-slip offsets.

Each of the above configurations are generally considered indicative of a

tensional origin. 39

Figure 26. Eastern Portion of the Crater Floor in Humboldt ■ ' 40 In addition, cracks are also observed to bisect minor structures, such as lower portions of crater walls (Figs. 17 and 26), small hills (Fig. 26), and the rims of small craters (Figs. 16 and 17) which lie with­ in cracked areas of the crater floors. In each case, the apparent lack of displacement along the crack argues against strike-slip faulting.

Cracks may also cross one another (Fig. 26); such configura­ tions would also be expected in structures of tensional origin. The termination of a crack as it encounters an earlier crack would only be anticipated if tensional stresses differed on both sides of the earlier fracture. Equally important, but less commonly observed crack configur­ ations are: 1. A braided pattern (Fig. 27) in which a crack splits, then recon-

verges to isolate a wedge of crater-floor material within the

crack,

2. A morphological transformation from a rille to a scarp (Fig. 28),

3. An increase in crack width, accompanied by the development •

of a wide, distinct floor (Fig. 29), and 4. • The association of cracks with crater chains and coalesced crater chains (Fig. 30).

Although these last four trace configurations are relatively rare, they establish geometric relationships which must eventually be explained.

Summary of Crack Descriptions

The occurrence of cracks is partially due to the size of the crater in which the cracks occur. Although larger craters may contain 41

Figure 27. Northeast Portion of the Crater Floor in Pitatus Zn i L )>

Figure 28. Crater Arzachel 43

Figure 29. Craters Repsold and Repsold G 44

Figure 30. Crater Messala 45 greater values of total crack length, the smaller craters often.have greater crack densities. This tends to suggest that either more energy is available for cracking within smaller craters or that the brittle fill­ ings within small craters are more susceptible to cracking.

Crack directions, in contrast, appear to be strongly related to the geometry of the craters in which the cracks occur, for most cracks show a distinct radial or concentric nature. Using the relative densities of radial and concentric crack elements,. it is possible to identify six major crack patterns. The regional control of crack direction is appar­ ently of minor importance, since only a small portion of the total number of observed cracks are aligned with the lunar tectonic grid. It is also observed that the cracks within craters are remarkably similar to the small-scale tension fractures produced in brittle materials.

As has been noted, terminating, crossing, branching, and braiding rela­ tionships are commonly encountered along the cracks.

In short, at least two important aspects of the cracks have been identified and need further investigation: the shallow cross-sectional profiles and the similarity of crack configurations to tensional fracture patterns. CHAPTER 3

CLASSIFICATION AND GENESIS OF LUNAR RILLES

Inspection of the high-re solution Lunar Orb iter IV photographs of the moon reveals the existence of several varieties of lunar rilles. Rilles can be classified by dimensions, configurations, and regional settings. For example, cracks were defined earlier in this study as rilles confined to the floor of a crater. In a similar manner, sinuous rilles receive their names from their characteristic serpentine traces .

Upon distinguishing rilles according to their morphology and setting, the problem of investigating the nature and origin of the cracks becomes manageable. Hopefully, it will be shown that rilles produced by fissuring differ from those produced by a flow of material within a channel or conduit. To contrast and compare the cracks with the other varieties,of lunar rilles, a general classification of lunar rilles is pro­ posed in this chapter. The order in which these classes of rilles are presented is: sinuous rilles, wide-floored rilles, irregular rilles, line­ ament rilles, coalesced crater chains, and cracks.

Sinuous Rilles

Many sinuous rilles resemble meandering terrestrial rivers be­ cause of their curved, winding channels. Typically, a sinuous rille is a single, unbranching feature whose sinuosity may be quite variable.

Quaide (1965) and Mutch (1970), for example, observed sinuous rilles

« 47 in the Aristarchus complex whose paths alternated between straight and sinuous segments. In other examples, such as seen in Figures 31 and

32, the sinuosity of a particular rille may be extremely pronouced with complex results.

Commonly associated with sinuous rilles are head craters or

irregular depressions located at the highest and widest end of the rille.

In both Cobra Head and Schroter's Valley (Fig. 33), the head crater is an oval depression that lacks the rim deposits normally associated with an impact crater.

A sinuous rille is also characterized by its nearly parallel walls. Mutch (1970, p. 195) observes that the cross-sectional profile of a sinuous rille uniformly decreases in both width and depth away from the head crater. The change in rille width might be quite slight over great distances; to the northeast of Gassendi an extremely long sinuous rille maintains a nearly uniform width of 0.6 km for a distance of 270 km.

The great majority of sinuous rilles, however, do not display such ex­ treme proportions. Only a few sinuous rilles exceed 100 km in length, and the mean rille width as derived from measurements of 21 sinuous rilles was 1.1km. The widest sinuous rille detected was Schroter1 s

Valley at 3.8 km.

According to R. Holcomb (1969, personal communication),

sinuous rilles are found along margins of maria and in suspected vol­ canic complexes, such as the Marius Hills. The upper end of the rille commonly originates either in the upland, on isolated topographic highs, or along the upland-mare contact, but the tail end of the rille almost without exception disappears on a mare surface, placing the formation of 48

Figure 31 . Crater Posidonius 49

upu upu , _ - \ mm ', ^ mm srl - a ~ o ,; ' ~ i - - ' 1

mm :##W ;

Ifm If m cp

Ifm If m mm

Posidonius mm srl

mm c cw

upu mm

upu frl

Figure 32. Map of Posidonius Derived from Figure 31 Figure 33. Schroter's Valley and the Cobra Head . 51 the rille as contemporaneous with the last stages of mare formation or even as post-mare in age. In addition, Fielder (1965, p. 115) and Mutch

(1970, p. 195) have observed that the path of a sinuous rille appears partially dictated by. the. existing topography; a sinuous rille usually , skirts hills or ridges that project above the mare surface . From these configurations, it may be argued that sinuous rilles are formed by the flow of material within a channel or conduit. The last diagnostic feature to be discussed is the gradation of . several sinuous rilles into chains of craters. The lack of rim deposits around these craters, their elongate shapes , and their uniform diameters closely resemble examples of terrestrial collapsed lava tubes. Accord­ ing to Hathaway (1971), terrestrial lava tubes occur in single flow units or several closely related flow units that have fused together. Molten lava exists beneath a chilled protective crust of the tube and starts to flow only when rupture occurs at a point of lower elevation in the flow.

Sinuous depressions will develop with the collapse of the protective ceiling immediately following withdrawal of the lava or at a later time when the roof has been weakened by seismic shocks.

Terrestrial lava tubes, however, rarely extend for more than several kilometers, while sinuous lunar rilles may exceed 100 km in length. A study of the Apollo 11 moon rocks by Tsutomu Murase and

McBirney (1970, p. 143) offers a solution to this disparity in length in their conclusions that the high iron and titanium content of the lunar basaltic rocks would significantly lower the viscosity of the original magma to the point where the lava would be far more fluid than any terrestrial lava. Extremely fluid lunar lavas would quickly form a chilled : ■ / - 52

crust under which they would flow for much longer distances in the lower

gravity of the moon. However, this model is jeopardized by the lack of identifiable collapsed roof fragments in most sinuous rilles. In contrast to the collapsed lava tube hypothesis, Firsoff (1961)

and Gilvarry (1960) have emphasized the similarity between the sinuous rilles and meandering terrestrial rivers and have argued for an actual

flow.of water on the lunar surface „ Cameron (1964) and Quaide (1965), on the other hand, oppose this hypothesis and favor the creation of sinuous rilles by glowing avalanche ash eruptions. Schumm (1970) pro­ poses still another hypothesis in which gas issuing along irregular fis­

sures beneath a nearly cohesionlesS debris mantle could create a linear zone of fluidized materials and, if these fluidized materials were to form on a slight slope, flowage along a channel could create a sinuous rille.

In summary, lunar sinuous rilles are structures produced by the

flow of volcanic materials at the lunar surface . The concentration of the sinuous rilles along the margins of the maria, in turn, identifies these regions as having been volcanically active.

Wide-floored Rilles

The wide-floored rilles have been known by many other names .

They have been referred to simply as rilles by Baldwin (1963) , Cruikshank

(1965), Ronca (1965), and Trask and Titley (1966), as clefts by Moore and

Cattermole (1967), as arcuate and linear rilles by Quaide (1965), as parallel-sided rilles by Firsoff (1961) and parallel-walled rilles by Gold

(1966), as straight rilles by Mutch (1970), as linears and lineaments by

Strom (1964) and Fielder (1965), and even as graben rilles, gaping 53 fissures, troughs, trenchesand rifts. To avoid this overlapping and confusing nomenclature, the new name of wide-floored rille is proposed here. The floor of a wide-floored rille may be either flat or hummocky, depending on the morphology of the surface structures crossed by the rille. If a wide-floored rille crosses the rim of a crater (Fig. 2), the crater rim appears to be downdropped into the rille. Wherever two wide- floored rilles cross one another (Figs. 2 and 34), the depth of the central area is approximately equal to the sum of the two rille depths. Along such rilles the inward-sloping rille walls maintain a nearly constant height. Such configurations have caused some workers (Baldwin, 1963;

Fielder, 1965; Quaide, 1965; and Kuiper, 1966) tentatively to identify these structures as grabens.

McGill (19 71) determined the wall slopes of the wide-floored rilles by measuring the change in rille width as a function of elevation.

In most examples, the rille slopes averaged 60°, and in one example a slope angle of 90° was discovered. McGill maintains that these slope values are suggestive of normal faulting and thus support the graben hypothesis. . z

Quaide (1965) separates the wide-floored rille s into two sub­ classes, corresponding to their traces; arcuate and linear. The arcuate wide-floored rilles are gently curved and commonly surround circular mare basins (Fig. 35), while the linear wide-floored rilles are associated with the irregular maria. Both subclasses display similar configurations

(Fig. 36) and consequently reflect similar origins. 54

Figure 34. Wide-floored Rilles and Strings of Secondary Impact Craters near the Crater Ramsden Figure 35. Wide-floored Rilles Adjacent to the Crater Campanus

Source: Orb iter IV photograph 131-1H

Scale: 1 cm equals 7.8 km

General location: Southeastern margin of Mare Humorum between the craters Campanus and Hippalus

The southeastern margin of Mare Humorum displays the best subparallel set of arcuate wide-floored rilles. Three prominent rilles of similar depth are present, plus a fourth shallow rille. The independence of the rilles from surface topography is clearly evident here as well as the typical graben-type morphology. North is toward the top of the photograph. Figure 35. Wide-floored Rilles Adjacent to the Crater Campanus 56

CONSTRICTIONS

EN ECHELON

INCOMPLETE BRAIDING

COMPLETE BRAIDING

Figure 36. Trace Configurations Characteristic of Wide-floored Rilles . 57 Measurements of 26 wide-floored rilles yielded a mean rille width of 2 , 6 km, thus establishing the wide-floored rilles as the widest type of lunar rilles. The greatest width measured was 8.5 km and the narrowest only 1.3 km.

Stratigraphically, the wide-floored rilles are often contempo­ raneous or slightly older than the youngest mare surfaces. The wide- floored rilles found along the eastern margin of Mare Serenitatis (Fig. 37) postdate the older and more heavily cratered mare surfaces but are, in turn, partially covered and filled by younger and darker mare mater­ ials. On the other hand, the wide-floored rilles encircling Mare Humorum

(Fig. 35) are younger than the mare surface they cross. Finally, crater chains are rarely associated with the wide- floored rilles. The notable exception to this rule is in the Hyginus Rille

(Fig. 38) where a series of large young craters cut into the steep rille walls. The origin most widely accepted is that floored rilles are graben structures related to the formation of the maria (Quaide, 1965, and

Mack in, 1969). Baldwin (1963, p. 377) proposes that the lava flows on the moon produce great domes and the lunar crust must bend under this weight to establish an isostatic balance. Furthermore, Baldwin suggests that the flexure near the mare margins might express itself as fissures

30 to 150 meters wide which are steeply inclined but not quite perpen­ dicular to the lunar surface. The head walls of these wide, steep fis­ sures would be expected to slump inward as lateral support is removed.

The observed insensitivity of the wide-floored rilles to the surface materials crossed (Figs. 34 and 35) implies that the surface topography Figure 37. Eastern Margin of Mare Serenitatis Figure 38. Crater Hyginus and the Hyginus Rille 60 is negligible in comparison to the thickness of the graben blocks.

The en echelon patterns found along the floored rilles creates some difficulties. Fielder (1965) and Moore and Cattermole (1967) at­ tribute the en echelon pattern to subcrustal strike-slip faulting. However,

Baldwin (1963) in reference to clay deformation experiments, points out that similar patterns could also result from slight rotation during sub­ sidence. The graben structure is challenged by several authors. Fielder

(1965) proposes a model in which upward-welling magmas associated with the initial fissuring melt their way upward into the frothy lunar surficial deposits, collapsing surface features along a line. These magmas would rarely reach the surface, but where they did, the linear features could be considered magmatic dikes (Ronca, 1965, and Moore. and Cattermole, 1967). Collapse features and hummocky zones within floored rilles would result from later subsidence of the magma and col­ lapse of the unsupported fracture walls. Baldwin (1963, p. 377) refutes

Fielder by calling attention to the great uniformity in width and depth of the floored rilles, regardless of the surrounding topography.

In short, the wide-floored rilles are extensive grabenlike fea­ tures whose distribution indicates zones of tension encircling the lunar maria.

Irregular Rilles

The irregular rilles are so named because of their complicated braided and branching behavior. Unlike the sinuous and wide-floored rilles, the irregular rilles are so interrelated that single rilles are .; ■ 6i difficult to delineate. Relationships commonly encountered in nets of irregular rilles are presented in Figure 39.

In general, irregular rilles and wide-floored rilles are just dif­

ferent expressions of the same linear feature (Fig. 40). The transforma­ tion of a wide-floored rille into an irregular rille is marked by a definite decrease in rille width along with the disappearance of a distinct floor unit. The change from a wide-floored rille to an irregular rille also . causes a change in depth and wall slope; the irregular rille is deeper

and has shallower wall slopes than its wide-floored counterpart. . . One of the most pronounced differences between the wide- floored and irregular rille is rille width. In contrast to a width of 2.6 km for the wide-floored rilles, the mean rille width derived for TO irregular rilles was only 1.1 km. The widest irregular rille measured was only

2.2 km across, and this value was obtained at a point just before the ■ irregular rille developed the characteristics' of a wide-floored rille. It is apparent from the rille-width measurements that the transformation from an irregular rille to a wide-floored rille is in some manner depen­ dent on rille width. Perhaps an irregular rille becomes'unstable when a critical rille width is reached. The irregular rilles also differ from the wide-floored rilles in regional distribution. Whereas wide-floored rilles commonly cross di­ verse surface structures, the irregular rilles are restricted to mare sur­ faces. In other words / the extent of the irregular rille is controlled by the nature of near-surface materials. It might be postulated that the ir­ regular and wide-floored rilles represent overlapping stress conditions; the wide-floored rilles may represent narrow zones of high tension above BR a R c H i //vg

/0//V(

cRo C u ? - r / /v,

r e/? ^//v ^//V (

e/v

/V T’r,ace 'O^r rat 06 Cfer i s t i'c re 63

!x

Figure 40. Wide-floored and Irregular Rilles near the Crater Triesnecker 64 deep fissures, while the irregular rilles are probably formed by more diffuse tensional stresses acting in brittle platelike materials. High concentrations of irregular rilles are generally found near the margins of the maria. Specific locations are near Galippus in the northeastern quadrant of Mare Serenitatis, in Lacus Mortis, extensively around Mare Orientale, to the southeast of the crater Hevelius , and along the eastern margin of Mare Serenitatis (Fig. 37).

Lastly, the irregular rilles are associated with a distinctive variety of crater chains. Unlike the sinuous rilles, in which the rille is transformed either into a sequence of craters or into a coalesced crater chain, the craters associated with irregular rilles are superimposed upon the rille in a manner similar to the craters situated within the

Hyginus Rille. In Figure 40 the craters found along two irregular rilles are approximately two to three times as wide as the rille in which they occur. On the whole, although the irregular rilles and wide-floored rilles share a common tensional genesis , they can be considered to be separate varieties of lunar rilles because of their distinct physical dif­ ferences and regional distributions.

Lineament Rilles

A lineament rille is an indistinct linear depression of highly variable character that occurs in the lunar uplands. Quite commonly, craters of contrasting size and morphology form the lineament rille, but wide, subdued valleys of obscure origin also account for many of their features (Fig. 41). Figure 41. Lineament Rilles near the Crater Airy 66 The restriction.of lineament rilles to the complexly shattered

and heavily cratered lunar uplands may explain this variable character.

The presence of the lunar tectonic grid system of major faults would naturally be expected to modify and control the effects of major asteroid impacts, regional tectonics, and pre-mare volcanism. Ejecta sheets from the impacts and volcanic eruptions of pyroclastics would also be expected, and these materials would tend to mantle and obscure earlier features. Therefore, it is quite possible that lineament rilles are simply major faults, wide-floored rilles, or even irregular rilles that have been modified by later tectonic and volcanic events. If this is true, then the morphological characteristics defining lineament rilles are primarily a result of age, not the original mode of formation.

In any event, the lineament rilles occur only on the lunar up­

lands and may be distinguished by their highly variable character from the other types of lunar rilles . f'

' Cracks

A comparison of the cracks with the sinuous, wide-floored, and irregular rilles (Table 1) shows that the cracks are essentially identical to the irregular rilles. Not only do the cracks display the same trace configurations observed among the irregular rilles, but they also show a fair agreement in terms, of rille width. In addition, the change in crack width as noted on the floor of Repsold (Fig. 24) is analogous to the transformation from an irregular rille to a wide-floored rille. The strong similarity between the cracks and the irregular rilles has two important consequences. First, this resemblance strengthens 67

Table 1. Characteristics of Four Major Lunar Rille Types

Sinuous Wide-floored Irregular Characteristic Rille s Rille s Rille s Cracks

Maximum width 3.8 km 8.5 km 2.2 km 2.5 km

Mean width 1.1 km 2.6 km 1.1 km 0 .7 km

Character single single multiple multiple

Trace sinuous, straight straight straight o c ca s. or to to straight arcuate irregular irregular

Uniformity constant or generally varie s from varies from slowly de­ constant, uniform to uniform to creasing but may irregular irregular width constrict

Cross- U or V generally, U or V U or V sectional shaped; a wide, flat shaped; shaped; profile may have floor; occas. shallow shallow narrow hummocky flat floor

Trace con­ unbranching crosscutting, crosscutting, crosscutting figuration branching, branching, branching, braiding, braiding, braiding, partial intercepting, intercepting braiding, en echelon en echelon

Sites originate in upland and restricted to restricted uplands at maria maria near to crater mare margins mare margins floors or in maria; terminate in maria

Special head crater, graben mor­ transitional radial and features rille within phology and with wide- . concentric rille wide floor floored rille s patterns

Crater observed rare observed observed chains 68 the tensional hypothesis of crack formation. Second, a parallel is noted between the crater-floor materials and the mare materials; in both struc­ tures, the cracks and irregular rilles are restricted to these materials and usually do not cut into the adjacent uplands. The question as to whether these two materials may be equated requires a closer look at the crater-floor materials. CHAPTER 4

CRATER-FLOOR MATERIALS

In general, the cracked materials of the floors of craters show great variety in morphology and albedo. Such diversity can only be at­ tributed to varying ages and compositions of the crater-floor materials.

Basic characteristics of these materials may be derived just from the mere presence of the cracks; in order to have a fracture, a medium must be coherent enough to be fractured. Furthermore, as is pointed out by Ronca (1966, p. 182) with respect to the cracks in Al- phonsus, "the regularity of some of the lineaments along their extent, especially the rilles, is an indication of the extreme areal homogeneity in the material which constitutes the floor of the crater." In a similar manner, other observations can be made from which the nature and the origin of the crater-floor material may be deduced. Such information is useful in suggesting mechanisms capable of causing or modifying the cracks. Consequently, Chapter 4 is an investigation of crater-floor materials associated with the cracks .

Original Crater-floor Materials

The first step in the investigation of crater-floor materials would be to examine the floor of the least altered, large impact crater.

Tycho with its extensive ray system would be the obvious example to study.

69 70 Shofthill and Saari (19.66, p, 216) scanned the lunar thermal disc in the 10 to 12 micron region during a total lunar eclipse and found

Tycho to be one of the strongest thermal anomalies on the lunar surface, representing a region of much slower cooling. Likewise, Hagfors (1966, p. 237) observed that Tycho reflects about ten times greater in the 68-cm wavelength than the immediately surrounding area. From this high reflec­ tivity it was inferred that Tycho is rougher than its surroundings and has a high intrinsic reflectivity. Both the thermal and radar reflectivity anomalies taken together have been interpreted as resulting from the presence of a relatively fresh, unaltered ejecta sheet surrounding the

Tycho impact site. Subsequent photographs taken by the Surveyor VII lunar probe which landed near the rim of Tycho have since confirmed the fragmental nature of the surface cover in this area.

The floor of Tycho, as shown in Figure 42, is characterized by its high albedo, intensely fractured morphology, and the occurrence of large, isolated hills; each of these features can be accounted for by a simple consideration of the physical processes operating during and im­ mediately following the impact explosion. The light color of the mater­ ials comprising the crater walls, floor fillings, and rays are probably due to shock lightening in which large numbers of internal fractures and cleavage surfaces are produced within mineral grains by the compressive shock wave. The " crackled and crenulated" surface texture of the floor materials (Mutch, 1970, p. 74), on the other hand, is limited to rela­ tively level areas between large hills and may result from irregular com­ paction settling within the coarse debris and crater-floor breccia. The origin of the large hills is more difficult to account for, but these Figure 42. Rough Floor Materials in the Central Portion of Tycho 72 structures may either be huge overturned blocks that were not ejected by the impact explosion or else represent shock-resistant structures, such as large igneous intrusions, that were exposed during the removal of the weaker overlying materials. A post-impact origin for these hills is un­ likely, for no small deposits suggestive of volcanism surround these

features to postdate the finely fractured floor filling.

The only smooth, flat materials in Tycho are not observed on the central portion of the crater floor but occur as light-colored deposits on slumped wall blocks and at the base of the crater walls. Cannon (19 71, personal communication) attributes these patches of level mater­ ials to mass-wastage processes, such as landslides, in which fine debris has moved down the crater walls.

In.general, the floors of craters which contain cracks do not resemble the rough fillings of fresh impact structures. Large hills simi­ lar in size and shape to those found in Tycho are commonly encountered, but the highly fractured surface morphology is missing. Since an impact origin is the most probable cause for these large lunar craters, it can only be concluded that either the original floor breccia has been modified in time by various geologic processes, such as meteoritic bombardment and post-impact volcanism, or else the early lunar materials were hotter and thus behaved more plastically than the cold materials in which Tycho formed.

Dark Secondary Crater-floor Fillings

Pre-Orb iter earth-based photography of the moon reveals sharp tonal differences of the materials found on the floors of craters containing cracks (Whitaker et a l., 1964), but the low resolution of these photographs does not permit accurate mapping of tonal changes.

The Lunar Orb iter IV photography, on the other hand, provides the neces­ sary resolution, but the low, variable lighting angle employed (180-23°) tends to emphasize changes in topography at the expense of albedo.

Since no standardized high-resolution maps of lunar albedo exist at -1 present, tonal changes will be referred to as the relative values of light and dark throughout this chapter.

Although the majority of the cracked floors within craters are relatively light colored, a few dark-floored craters, such as Camp anus

(Fig. 35), Pitatus (Figs. 13 and 14), Hesiodus (Figs. 13 and 14), and Tsiolkovsky (Fig. 43) also exist. In many other examples, dark floor materials only occur as small patches located in a variety of topographic and stratigraphic positions.

In Tsiolkovsky (Fig. 43), for example, the extremely dark floor materials fill the central portion of the crater floor but are encircled by a narrow strip of rugged, light-colored floor materials that lies at the base of the crater walls. Along the ragged but sharp tonal contact be­ tween the light and dark materials, the dark fillings occasionally sur­ round isolated mounds and ridges of the lighter material which is all that remains of the nearly completely buried original floor breccia . The extreme levelness of this dark filling may indicate its emplacement as a highly fluid laval, while the narrow cracks at the outer margin of the dark filling postdate the consolidation of the lava, and may indicate com­ paction of the floor breccia by the lava filling. 74

Figure 43. A Close-up View of the Northern Portion of the Crater Floor in Tsiolkovsky 75 In Petavius (Fig. 44) the dark floor materials are found in the low-lying region at the base of the domed floor of the crater. Two promi­ nent cracklike rilles terminate in the arcuate patch of dark materials, but upon closer inspection it can be seen that both of these rilles display several features characteristic of the sinuous rilles. All turns and bends along these rilles are gently curved as if carved by an erosive flow of material, large depressions resembling head craters are found near the upper ends of these rilles, and the rille widths gradually decrease away from the high central portion of the crater floor. The branching off of smaller cracks from these two rilles, however, is not typical of sinuous rilles . What has probably occurred is that both rilles originated as radial cracks when the crater floor was updomed and subsequent volcanic eruptions from the central portion of the crater floor have utilized the channels provided by the cracks to reach the low moat that surrounds the domed floor. Materials flowing within these channels were probably dense clouds of flowing ash instead of lavas because lavas would tend to be restricted to their channels and thus could not darken the hummocky regions adjacent to the channels .

In Alphonsus (Figs. 23, 24, and 45) a third variety of dark crater-floor material occurs as dark halos surrounding craters that are superimposed upon cracks. In many places, these dark halos extend up to several kilometers from the source crater and then gradually blend in with the light-colored floor materials without showing any significant changes in crater density. Whitaker (1966) confirms the thin veneerlike nature of these dark halos by the observation within one of the craters of a line of dark spots just below the crater rim with talus deposits 76

Figure 44. Modified Cracks and Associated Dark Floor Materials Within the Crater Petavius Figure 45. Dark-halo Craters on the Western Portion of the Floor of Alphonsus

Source: Orb iter V photograph 118-M

Scale: 1 cm equals 1.2 km General location: Northwestern margin of Mare Nubium

Coordinates: Lat 13.5 S. , long 2 . 7 W .

Diameter: 114 km

The rim deposits of the large, elongate dark-halo crater can be seen covering and filling the crack that intersects the north side of this crater. The rim deposits are relatively smooth and not hummocky and therefore suggest a volcanic origin. The abundance of soft-rimmed, shallow crater chains and indistinct major lineaments that cover the entire area surrounding these dark-halo craters indicates the strong link between lunar volcanism and tectonics.

North is toward the top of the photograph. 77

Figure 45. Dark-halo Craters on the Western Portion of the Floor of Alphonsus 78 descending from each of the spots. All of the above observations, plus the general uniformity of surface texture, the symmetrical distribution of dark materials around a dark-halo crater, and the presence of old, un­ filled small craters within the dark halos, argue for explosive, gas- charged eruptions of dark pyroclasties. Workers who have developed this picture for the formation of the dark-halo craters in Alphonsus are

Carr (1966, p.: 275), Gold (1966, p. 119), Kuiper (1966, p. 102,

McCauley (1966, p. 316), Mutch (1970, p. 230), O'Keefe (1966, p.

261) > Schumm (1970, p. 2539-2541), Smith (1966, p. 255-256), Whitaker

(1966, p, 93), and Urey (1966, p. 9). Gold (1966, p. 118), however, has also proposed that the dark halos are due to a discoloration produced through a reaction of lunar surface materials with chemicals exhaled from the interior of the moon.

The various dark materials just examined in Tsiolkovsky,

Petavius, and Alphonsus demonstrate that all dark crater-floor materials are not identical but may represent a complex series of volcanic products ranging from lavas to gas-charged pyroclastics.

The relative ages of these dark crater-floor materials may also vary. For example, although the cracks in Tsiolkovsky (Fig. 43) are younger than the dark materials in which they occur, the situation is reversed in many other craters in which the dark crater-floor materials cover older cracks. The dark materials in Humboldt (Figs. 18 and 19) and Schluter (Fig. 20) fall into this latter category.

In short, it may be inferred from these few examples of dark crater fillings that the modification of a crater floor is a complex pro­ cess, often volcanic in nature, that differs from crater to crater. 79 Additional examples of the various occurrence of dark materials found on crater floors appear in Table 2.

Table 2. Examples of Dark Crater-floor Materials Found in Craters Which Contain Cracks

Type of Dark Filling Examples

Entire floor of crater is dark Campanus Pitatus Hesiodus Kopf

Dark, level fillings that predate Tsiolkovsky cracks Campanus Boscovitch Pitatus Hesiodus Kopf

Dark, level fillings that postdate Humboldt cracks Hansteen Schltiter Einstein "V" Group of craters surround­ ing Mare Smythii

Craters containing dark-halo craters Alphonsus CleomedeS Schrodinger Franklin

Irregular patches of dark materials Messala La Condamine Petavius

Light-colored Secondary Crater Fillings

The crater-floor material most commonly associated with cracks

is a light-colored filling that varies in morphology from level to hum-

moacky surfaces. Stratigraphically, this light-colored material always

occurs above the rough floor breccia but below the dark crater-floor materials; in no instance are light-colored materials ever observed to overlap the dark fillings. Wherever both light and dark secondary fill­ ings are in contact, the higher density of large craters on the lighter materials indicates a greater age for the light-colored filling. These relationships are consistently observed over the entire lunar surface and may be the result of two separate phases of lunar volcanism „

In the case of Alphonsus, many authors (Carr, 1966; McCauley,

1966; O'Keefe, 1966; and Ronca, 1966) generally agree on a pyroclastic origin for the light-colored crater-floor materials. The widespread occur­ rence of shallow, circular depressions (ghost craters) on the floor of

Alphonsus and the subdued morphology of cracks (Fig. 45) are the mor­ phological features one would expect if the crater floor had been covered by pyroclastic debris. According to O'Keefe (1966,„ p. 185), Mackin.

(1969, p. 744), and Mutch (1970, p. 185), a low dense cloud of hot ash would settle as it moves over an area, compacting later to form a muted version of the underlying surface; small craters would disappear in the process, while larger craters would become shallow depressions follow­ ing differential compaction.

The most likely eruptive sites for the light-colored ashlike deposits are the cracks. McCauley (1966), in mapping a portion of the floor of Alphonsus from Ranger IX photographs, was able to delineate several distinct, overlapping floor units of differing crater densities that were symmetrically distributed around the cracks. Carr (1966) in a separate study of Alphonsus derived similar results. Figure 46, in which two light-colored, low-rimmed craters in Alphonsus are shown situated directly over cracks, also suggests that volcanic materials have erupted 81

Figure 46. Thick-rimmed, Light-colored Craters along Cracks in Alphonsus 82 from the cracks. Similarly, cracks have apparently also served as erup­ tive vents for ignimbritic flows in Gassendi, because in Figure 47 a zone of shallow, partially filled craters is observed midway between the cracks and the highly cratered central portion of the crater floor; this transitional zone probably marks the outer margin of the ash flows. The floor of Petavius (Pig. 48) also displays subdued and partially filled cracks and large craters, indicating similar processes operating within this crater. An ignimbritic origin for the light-colored deposits in craters containing cracks would explain several of the characteristic features of the cracks but would leave other features still unexplained. The low slopes observed on the walls of the cracks, the lack of small sharp craters and rock outcrops, and the "tree-bark" surface texture all agree with the presence of a fine-grained, unconsolidated pyroclastic debris layer that has been compacted by gravitational settling. It would seem highly unlikely, however, that such poorly consolidated materials would be capable of developing wide fractures.

A brittle material must be proposed which underlies the poorly consolidated, light-colored surficial layer to account for the tensional nature of the cracks. Several possible causes exist for this layer:

1. Most of the crater filling was erupted at a single time in a

massive outpouring of ignimbritic materials. Residual heat of

these materials and gravitational compaction then welded the

bottom of the sheet to form a brittle layer. 83

Figure 47. The Central Section of the Floor of Gassendi Figure 48. Hummocky Floor Materials within the Crater Petavius

Source: Orb iter V photograph 35-M

Scale: 1 cm equals 2.1 km

General location: Southwestern margin of Mare Fecunditatis

Coordinates: Lat 25.3 S . , long 60.4 E

Diameter: 183 km

The hummocky materials in Petavius are marked by several characteristics indicative of a complex volcanic filling: variable albedos, discrepancies in density of cratering, an abundance of small dome-shaped hills, and a near obliteration of several cracks as a result of filling.

North is toward the top of the photograph. Figure 48. Hummocky Floor Materials Within the Crater Petavius 85 2. An old.lava filling, created either by a single outpouring of lava

or by a sequence of flows in which the brittle filling was grad­ ually built up, niay underlie the younger unconsolidated layer. 3. The impact crater may have formed in a much hotter state of the moon's development. Excavation by the impact explosion

may have exposed submolten materials which then welded upon

cooling to produce a brittle crater floor quite unlike the crater- floor breccia observed in Tycho (Fig. 42) .

Small-scale Features of Possible Volcanic Origin

There are many small-scale features on the cracked floors of lunar craters, such as domes with central pits, double-ringed craters, and crater chains, whose morphology and mode of occurrence indicate a volcanic origin. Such small-scale features are important in that they help to establish the role of volcanism in the filling of the crater floors.

Domes with central pits are one of the closest lunar analogies to volcanic craters. These features closely resemble normal impact craters and can only be distinguished by convex profiles on the flank slopes, by the occurrence of crater floors at a higher elevation than the surrounding topography, and by a rim deposit that is greater in volume than the central pit. Such features are commonly only a couple of kilo­ meters in diameter and may be superimposed upon a crack or along a crater chain. Such well-defined features are relatively rare but are present in Lavoisier D (Fig. 10), Schluter (Fig. 20), Arzachel (Fig. 28), Alphonsus (Figs. 23 and 24), Hesiodus (Fig. 13), and Gassendi. 86 Many of the hill-like domes found on the cracked floors of lunar craters may actually be volcanic in origin, but at present these cannot

be distinguished from the large h ills, like those observed in Tycho

(Fig. 42), which were apparently created during the initial impact ex­

plosion. An additional problem can be created by the coincidental impact

of a small meteorite at the top of one of these non-volcanic hills; the

resultant feature would resemble a dome with a central pit. Therefore,

a safe approach in identifying possible volcanic structures would be to rely upon the occurrence of the suspected features along a linear struc­ ture, such as a crack, in which structural control of volcanism may be

inferred. The chain of conelike craters within Hesiodus (Figs. 49 and

50) were identified in this manner because they are situated at the inter­ section of a crack with a well-defined structural lineament.

One of the most unusual features associated with lunar craters with cracked floors are the double-ringed craters. These features con­

sist of a small crater situated directly over the center of the larger

crater; the diameter of the small crater is generally about one-half the

diameter of the large crater. The rim crests of both craters are usually similar in sharpness, implying similar ages, while the rim deposits of

both craters have smooth slopes, indicating similar compositions. A

coincidental sequence of meteoritic impacts can be ruled out by the lack

of typical impact features, such as hummocky rim deposits, rays, and

secondary impact craters, and by the exact centering of the smaller

crater within the larger one. Of the ten double-ringed lunar craters so

far identified by the author, five occur within craters that have cracked

floors (Lavoisier, Fig. 15; Humboldt, Fig. 26; Repsold, Fig. 29; 87

Figure 49. Eastern Portion of the Floor of Hesiodus Figure 50. Map of Eastern Portion of Hesiodus Derived from Figure 49

Key to Map Symbols cch crater chain ic indistinct crater ccw collapsed crater wall mm mare material dfm dark floor material sc subdued crater fc fresh crater 88

sc \\

ccw _z - \-

IC

d f m

//_ \ i

_\y

- \ \ x \ : z/. X > - / sc

1C

ccw // ccw T ^ -' - / dfm )/ -

sc cch

ccw dfm 7/

sc dfm \

ccw

ic ic

Figure 50. Map of Eastern Portion of Hesiodus Derived from Figure 49 . 89 Mersenius, Figs. 51 and 52; and Cruger E), three are adjacent or near

craters containing cracks (Hesiodus, Figs., 13 and 14; Einstein; and Schrod.inger), and the remaining two are situated over prominent linea­

ments. The smooth-rim deposits, the close association between the

double-ringed craters and tectonic linear features, such as cracks and

lineaments, and the similar rim profiles of both the large and small

craters tend to suggest a two-phased, gas-charged, set of explosive

eruptions from the same vent with the second one being of lower inten­ sity. The last small-scale feature to be discussed is the crater •

chains. The widespread occurrence .of small craters of similar size and

cross-sectional profile are too statistically improbable to be attributed to a random process, such as meteoritic impacts; such features must be the product of lineament-controlled volcanism or subsurface drainage.

Kuiper (1966, p. 100) in regard to these crater chains suggests that they are produced by the escape of gases from a cooling and shrinking magma body. However, other workers, like Carr (1966, p. 221), Gold (1966, p. 118), McCauley (1966, p. 317), Urey (1966, p. 8, 15, and 19), and

Whitaker (1966, p. 93) favor collapse and drainage of unconsolidated materials following fissure formation. Degassing would tend to produce craters with well-formed rim deposits like the chain of large craters in

Mersenius (Figs. 51 and 52), while drainage, on the other hand, would create low-rimmed, soft-edged craters of conical cross-section, such as are commonly observed in Alphonsus. Other craters with cracked floors that display prominent crater chains are Einstein A, Hevelius,

Lavoisier D, and Messala. 90

Figure 51. Crater Mersenius Figure 52. Map of Mersenius Derived from Figure 51

Key to Map Symbols cch crater chain hmk hummocky floor ccw collapsed crater wall material cp central peak cw crater wall Ifm light-colored floor material dm dome drc double-ringed crater sc subdued crater fc fresh crater upu undifferentiated frl wide-floored rille upland material 91

id m? upu If m If m

'cp drc

ccw ccw Ifm

cch cch

Ifm

Mersenius

url

sc

upu frl

Figure 52. Map of Mersenius Derived from Figure 51 92 Summary of Crater-floor Materials

The development of the floor filling within a lunar crater can be

the result of many diverse processes . If the original crater were created by the impact of a small asteroid body on the lunar surface, then a rough

floor breccia surrounding large, hill-like blocks might be normally ex­

pected. However, if the impact had occurred during a period of high heat flow, excavation of the cooler overlying rocks might expose a zone

of submolten material which could flow into the crater and weld upon

cooling. Geologic processes that could then modify and cover this

original crater floor would be meteoritic bombardment, talus flows from the crater walls, mantling by sheets of impact ejecta from nearby major impacts, and post-impact volcanism.

As observed in this chapter, post-impact volcanism may occur

in a variety of forms, and the floor of a crater may contain materials

from several periods of volcanism. The filling most commonly encoun­

tered is a relatively light-colored material that suggests emplacement by

ignimbritic eruptions . This material subdues earlier features and appears

to be poorly consolidated. However, the existence of a brittle material

underlying this poorly consolidated layer must be postulated to explain the tensional pattern of the cracks. This brittle material may either be

a welded portion of the overlying pyroclaStic debris or an earlier lava

filling. Postdating the light-colored surficial material there are several

varieties of dark volcanics that range from lava fillings to small, pyro-

clastic blankets. The occurrence of diverse, small-scale volcanic fea­

tures on the floors of craters containing cracks also implies a complex

volcanic history for the filling of a crater floor. CHAPTER 5

CRATER FLOOR AND RIM MORPHOLOGY

A large portion of the geologic history of a crater with cracks on its floor may be deduced from the general morphology of its floor and rim. Earlier considerations in this thesis have shown that the floors of many lunar craters have been filled with a level, brittle material of coherent nature. This material, when subjected to unequal vertical forces, would act as a thin brittle plate and would tend to deform in a manner reflecting the geometry of these forces. Within this chapter several possible mechanisms for the deformation of the floor plate are deduced from the crack patterns and floor profiles. Crater-rim struc­ tures are also contrasted in terms of meteoritic impact and volcanic mechanisms of formation. Lastly, the age of the crater is discussed from the' point of view of the morphology of the crater rim.

Crater-floor Morphology

The brittle floor plate, in effect, records the crater's tectonic history after the initial stage of formation. Crater-floor profiles, inferred from photographs, and crack patterns suggest the mode of post-formation deformation of the craters. Associated with each stage of tectonic defor­ mation is a particular fracture pattern. For example, vertical uplift pro­ duces a radial fracture pattern, as commonly observed over terrestrial

salt domes. Subsidence, on the other hand, results in a network of concentric fracture. Keeping in mind these characteristic fracture

93 94 patterns, the general crater-floor profile gives additional clues as to various stages of the crater's history. Examples of commonly encoun­ tered profiles include: (a) convex floor, (b) convex floor with central

depression, (c) irregular floor, (d) concave floor, and (e) flat floor

(refer to Table 3 for specific examples).

Table 3. Examples of Various Crater-floor Profiles of Craters Containing Cracks

Crater-floor Profile Craters

Domed Einstein "V" Humboldt Lavoisier F Lick Mersenius Petavius Repsold Partially collapsed; irregular Arzachel Damoiseau Gassendi Hansteen La Condamine Messala Schluter Vitello Concave Briggs Gauss Hesiodus Lavoisier Pitatus Ritter Approximately level Alphonsus Atlas Cardanus Cleomedes.

One series of tectonic events that could explain several of

these profiles starts with the formation of the brittle floor plate (refer 95 to the previous chapter), followed by an injection of molten material beneath this plate. Assuming a relatively thin floor plate and a mobile magma, a domed structure would form, which 1. Could freeze in place as illustrated in Figure 53a and represent­

ing Lavoisier F (Fig. 15); 2. Allow some of the molten material to seep out as it freezes

(Einstein "V", Figure 22);

3. Start to collapse in sections as illustrated in Figure 53b

(Hansteen, Fig. 54); or

4. Completely collapsed (La Condamine, Fig. 55).

The initial uplift in this sequence creates a radial fracture pattern, while collapse of the dome results in a superposition of concentric fractures.

In contrast, a thick floor plate, rather than doming, would be uplifted as an intact unit, as shown in Figure 56b and represented by the craters

Posidonius (Figs. 31 and 32) and "Mare Smythii D" (Fig. 57) . This would result in few fractures in the plate but would create a moatlike zone sur­ rounding the floor plate.

Another model starts with the development of a magma beneath a flat floor plate followed by subsidence of the floor as the magma is withdrawn or shrinks upon cooling. This process is illustrated in Figure

56a and may be responsible for the development of peripheral Concentric cracks and concave floor profile of Pitatus (Fig. 27).

A third model consists of a flooding of the crater by a fluid lava, followed by concentric cracking along the periphery of the lava filling due to compaction of the underlying impact breccia, giving a flat UPDOMING

w

UPDOMING AND SUBSIDENCE

Figure 53. Updoming and Partial Collapse of the Brittle Floor Plate Figure 54. Crater Hansteen

Source: Orbiter IV photograph 149-2H

Scale: 1 cm equals 5 . 2 km

General location: Northwest of Gassendi along the western margin of Oceanus Procellarum

Coordinates: Lat 11.5 S ., long 51.9 W.

Diameter: 46.0 km

The floor of Hansteen appears to be domed and intensely cracked. Portions of the light-colored floor material are missing and presumed collapsed. The material filling these collapsed areas is very smooth, dark, and lightly cratered and is nearly identical to the mare material overlapping the base of the crater rim deposits.

North is toward the top of the photograph. Figure 54. Crater Hansteen Figure 55. Crater La Condamine

Source: Orbiter IV photograph 145-1H

Scale: 1 cm equals 3.7 km

General location: In the uplands north of Sinus Iridum

Coordinates: Lat 53.4 N. , long 28.1 W.

Diameter: 37.7 km

The central portion of the floor of La Condamine is a depression which may be due to collapse of the centralmost sec­ tion of a domed floor p late. This hypothesis is suggested by the raised ring on the crater floor that encircles the central depres­ sion. Bounding this depression are steep scarps along which the upper portion of a dome is inferred to have collapsed. Dark materials within the central depression also suggest that this portion of the crater floor has developed differently than the light-colored floor ring.

North is toward the top of the photograph. Figure 55. Crater La Condamine SUBSIDENCE

UPLIFT OF PLATE

m m

Figure 56. Uplift of a Thick Brittle Plate and Subsidence of a Brittle Plate Figure 57. Crater "Mare Smythii D"

Source: Orb iter IV photograph 017-3H Scale: 1 cm equals 3.2 km

General location: Along the margin of Mare Smythii in the eastern limb region of the moon

Coordinates: Lat 1.0 S. , long 84.5 E. (ap­ proximate)

Diameter: 38.7 km (approximate)

Note the upraised central section of the crater floor and the mare material flooding the low-lying zone that encircles this central plate. North is toward the top of the photograph. Figure 57. Crater "Mare Smythii D" 101 to slightly concave floor profile. The narrow cracks in Tsiolkovsky (refer to chapter 4) may have been formed in this manner. A fourth model, also resulting in peripheral concentric fracturing and a level to slightly con­ cave floor, includes tectonic reactivation of buried crater-wall subsi­ dence blocks as a result of lubrication by the magma. Furthermore, this model explains those cracks that continue from the crater floor into the crater walls.

Lastly, a fifth model begins with the rejuvenation of regional faults independent of any volcanism. This process could account for the few cracks observed to be oriented parallel to the lunar tectonic grid

(refer to chapter 2, Figs. 8 and 10 of Hevelius and Lavoisier D, respec­ tively). In summary, it appears that many tectonic mechanisms are responsible for the morphology of the crater floors. However, as shown in this discussion, a somewhat systematic description of possible se­ quences of tectonic activity can be developed from morphological fea­ tures. In any event, these five models should not be considered the only processes by which crapks can be created.

Crater-rim Morphology

The rim deposits of only a few craters with cracked floors show definite signs of an impact origin. Cardanus (Fig. 58) is probably the best example among the cracked craters of an imp act-produce d crater, displaying a high, hummocky rim with faint radial depressions. In sup­ port of the impact mode of formation of Cardanus are strings of secondary Figure 58. Crater Cardanus 103 Impact sites in the surrounding mare m a te ria l.B rig g s (Fig. 59) also

shows a hummocky rim, indicating an impact origin, but is postdated by the surrounding mare materials.

In other examples, a few craters with cracked floors display possible volcanic or volcanically modified rims. These rims are charac­ terized by features such as:

1. Little or insignificant rim deposits (Kopf, Fig. 60; Schluter "X" ,

Fig. 61; and Damoiseau, Fig. 62), and

2. Low, smooth rim deposits which may be the result of volcanic mantling (the Lavoisier group of craters. Fig. 1).

In general, however, only a few craters with cracked floors have definite characteristics that indicate a specific mode of origin. The majority of craters with cracked floors have highly modified rims, the origins of which are not immediately apparent, such as Helvelius (Figs. 8 and 9),

Lavoisier D (Fig. 10), Pitatus (Fig. 13), Alphonsus (Fig. 23), and La

Condamine (Fig. 55). Pohn and Offield (1969) have interpreted highly modified crater rims as indicating extremely ancient craters. The above-noted proba­ bility of volcanism associated with craters with cracked floors implies that highly modified rims may not always be indicative of very old craters . Therefore, the Pohn and Offield age criterion may not be appli­

cable to craters with cracked floors.

1. To the northeast of Cardanus, an unusually smooth, dark patch of mare material covers'some of these secondary impact craters, suggesting that Cardanus is contemporaneous with the formation of the mare. 104

‘ t:: Figure 59. Crater Briggs 105

Figure 60. Crater Kopff Figure 60. Crater Kopff

Source: Orbiter IV photograph 187-2H

Scale: 1 cm equals 4.4 km

General location: Northeast margin of the inner basin of Mare Orientale

Coordinates: Lat 17.5 S. , long 89.0 W .

Diameter: 41.3 km

Kopff has sm all, smooth rim deposits that may be con­ sidered negligible. The crater had to have formed after the event that formed Mare Orientale because of the sharp rim crest and the steep scarplike walls. Another large post-basin formation crater near Kopff # Maunder, has a hummocky rim and well-defined ray system and is in sharp contrast to Kopff.

The floor of Kopff is very dark and very smooth and identi­ cal to the materials filling the inner basin of Mare Orientale. The cracks within Kopff are sharper rimmed than those for other lunar craters with cracked floors, implying less modification, but the crack widths are approximately equal to those in other craters with cracked floors. North is toward the top of the photograph. Figure 61. Crater Schluter "X"

Source: Orbiter IV photograph 187-1H

Scale: 1 cm equals

General location: In the Orientale ejecta blanket ap­ proximately 300 km northwest of crater Schluter

Coordinates: lat IS., long 90 W. (approximate)

Diameter: 12.9 km

Schluter "X" must be younger than the Orientale ejecta blanket because no trace of the blanket can be seen within the crater. It is also highly improbable that Schluter "X" is an impact structure because the rim deposits are low and smooth. In addition, the morphology of the crater-floor materials are quite unlike any­ thing else in this vicinity and may be due to a volcanic filling.

North is toward the top of the photograph. Figure 61. Crater Schluter "X" Figure 62. Crater Damoiseau

Source: Orb iter IV photograph 161-1H

Scale: 1 cm equals 5.5 km

General location: Western margin of Oceanus Procellarum near the crater Grimaldi

Coordinates: Lat 4.9 S . , long 61.1 W .

Diameter: 36.5 km

The old large crater in which Damoiseau occurs in a off- centered position has a distinct, though subdued, rim. Damoiseau, in contrast, appears to be a much younger structure with insignifi­ cant rim deposits. From the sharpness of the rim crest and the lack of wall slump blocks, a non-impact origin is proposed for Damoiseau, possibly volcanic collapse.

North is toward the top of the photograph. 107

Figure 62. Crater Damoiseau 108 In conclusion, except in a few cases, the rims of craters with cracked floors seem to disclose little of the geologic history of these craters. The morphology of the crater floors, in contrast, may reveal much of the past deformational events affecting these craters.

\ CHAPTER 6

REGIONAL CONTROL OF CRATERS WITH CRACKED FLOORS

In general, cracks are confined to crater floors in either sub- radial or subconcentric patterns, implying the existence of a crater- centered structural control for these features„ However, only a small portion of the total number of major lunar craters have cracked floors.

To explain this restricted occurrence of cracking, a higher level of structural control, perhaps one on a regional basis, might be expected.

Chapter 6, therefore, is an investigation of the regional distribution of the cracked craters.

Size-frequency Distribution of Lunar Craters with Cracked Floors The class of lunar craters with cracked floors has a distinctive size-frequency distribution. Figure 63 shows this distribution for the lunar frontside.

Extremely large cracked craters like Humboldt (202 km),

Petavius (183 km). Gauss (174 km), and Alphonsus (114 km) determine the upper limit in size for the lunar frontside distribution, while craters such as Schrodinger (230 km), Oppenheimer (220 km), and Tsiolkovsky

(110 km) establish the upper limit for the lunar far side. These large circular structures are intermediate in size between the largest craters and the circular mare basins and often show structures common to both the craters and basins. Wilhelms et al. (1965), for example, correlated

109 rtr wt Cakd Floors Cracked with Craters

NUMBER OF CRATERS 0 1 15 -- -• iue 3 LnrFotie iefeuny itiuin of Distribution Size-frequency Frontside Lunar 63. Figure 20 40 60 RTR IMTR (KM) DIAMETER CRATER 100 120 14060 160 110 Ill structures in the rim materials of Petavius with those surrounding Mare Crisium. In addition, it should be recalled that several of the circular maria are encircled by irregular and floored rilles. Properly speaking, such circular maria should also be included in the size-frequency dis­ tribution of craters with cracked floors, since the cracks, irregular and wide-floored rilles are all interrelated. If the large craters, like Hum­ boldt and Oppenheimer, differ from the circular maria only in size and degree of filling, then a possible genetic link might exist between the cracks and the mare filling of the circular basins. Between 30 and 200 km, the size-frequency distribution of craters with cracked floors is nearly exponential, but between 9 and

30 km, the crater frequency drops off sharply and a distinct lower cutoff occurs at a diameter of 9 km . The exponential portion of this frequency distribution might be related to the exponential size-frequency distribu­ tion of craters produced by meteoritic impact. The drop in frequency and lower size limit might then represent the increasing inability of lower energy impacts to breach a critical, subcrustal lunar layer. The observation of slumped rims on lunar craters whose diam­ eters are greater than 20+5 km caused Mackin (1969) to postulate a subcrustal zone of high temperature that was in existence at the time of meteoritic impact. Immediately following impact and excavation, the relief of. lithostatic pressure through the sudden removal of lateral sup­ port was sufficient to cause material within this zone to melt. The combined loss of lateral support and lessened viscosity would then favor inward slumping of the crater walls soon after impact. Only those impacts capable of producing craters at least 15 km wide were able to 112 upset this critical subcrustal layer. Depending on temperatures within this critical zone, one might also expect a molten filling to form in the central portion of the excavation. As a consequence of Mackin's model, both volcanism and subsequent tectonism are initiated by impact. An important point to remember is that not every crater greater than 15 km in diameter has a cracked floor nor does every crater over this size have slumped walls. The extent of areas underlain by a high- temperature zone was probably limited, and the depths and durations of particular zones were probably variable. In fact, the formation of a high- temperature zone might have occurred long after the impact explosion, and a regional increase in heat flow could cause wall slumpage through overall loss of the supportive strength of crustal materials. If the crack­ ing of crater floors is due to a volcanic-tectonic process, a regional separation of craters with cracked floors might delineate ancient regions of high heat flow in which older meteoritic impact craters were modified.

Distribution of Craters with Cracked Floors by Latitude and Longitude

One method of studying the distribution of craters with cracked floors is to sort them by latitude and longitude. If the distribution of these craters is random, then the data should be nearly independent of latitude and longitude.

Distribution by Latitude

Lunar frontside craters with cracked floors were initially sorted into 10° intervals from north to south. However, these unmodified data are misleading because the 80°-90o north latitude belt covers a much 113 smaller area than the equatorial 0°-10o north latitude belt. An uncor­ rected graph of crater frequency versus latitude would naturally favor the equatorial regions because of the larger areas included within this zone. To compensate for this discrepancy in area, the area of each belt was calculated and a corresponding Correction was derived, which when multiplied by the area within the belt would yield a standardized area.

The calculations used in deriving these factors are presented in Figure

64: The corrected frontside crater-frequency distribution by latitude appears in Figure 65.

In spite of the corrections, the greatest occurrence of craters with cracked floors is within 60° of the equator, basically the same distribution as the maria. This similarity also strengthens the possi­ bility of a genetic link between the maria and the cracked craters. In addition, the maximum frequency distribution of frontside craters with cracked floors is slightly skewed toward the northern lunar hemisphere along with the distribution of the lunar maria.

Distribution by Longitude

Figure 66 displays the distribution of frontside craters with cracked floors by longitude. The data are nearly random with the excep­ tion of a strong concentration of craters along the western limb region, coincident with the western margin of Oceanus Procellarum. Data for the eastern lunar limb region are incomplete due to poor Lunar Orb iter IV photographic coverage of this area. 114

R d0

0 = CO-LATITUDE ANGLE

A = AREA OF BELT

A = (2TTD) RdO

D = R sin 0

A = 2 7T R2 sinO d 0

CALCULATION OF AREA CONTAINED WITHIN BELT FROM 0, TO 0O

A = 2 7TR sin 0 d 0

0 , A = — 2 7T R ^ cos 0 = K ( cos 0 2 — cos 01 ) 0

AREAL CORRECTION FACTORS :

(9 0 ’ — 8 0 ’) » 13.2 (40' -3 0 ) 1.40

(80 - 7 0 ' ) = 4 .4 3 (30 -20') 1.27

(70" — 6 0 ’) = 2.73 (20' -10) 1.19

(60 - 5 0 ) = 2 .0 0 ( 10 -O') = 1.15

(5 0 ’- 4 0 ) 1.62

Figure 64. Calculations for Standardizing Areas within 10° Latitude Belts 90 80 70 60 50 40 30 20 10 0 10 20 30 40 50 60 70 80 90 S0UTH LATITUDE - 10° UNITS NORTH

Figure 65. Corrected Latitude Distribution of Lunar Frontside Craters with Cracked Floors ih rce Floors Cracked with NUMBER OF CRATERS 22 26- 20 24- 10 12 14- 18- 16- 4- 8 - - - - - WEST Figure 66. Longitude Distribution of Lunar Frontside Craters Craters Frontside Lunar of Distribution Longitude 66. Figure a u OG TD - 1* UNITS 10*LONGITUDE - photographic incomplete coverage ein of region

EAST 117 Distribution of Craters with Cracked Floors in Relation to the Maria Figure 67 (in pocket) shows the distribution of frontside craters with cracked floors. A strong relationship is immediately observed; the majority of these craters occur along the margins of the maria or within irregularly bounded regions of shallow mare flooding. Table 4 lists prominent craters with cracked floors which are associated with the major maria.

Table 4. Examples of Craters with Cracked Floors That Are Associated with the Major Lunar Maria

Mare Craters

Crisium Cleomedes, Lick, Yerkes Fecunditatis Goclenius, Petavius, Taruntius Frigoris La Condamine Humorum Doppelmayer, Gassendi, Vitello Imbrium Cassini Nubium Alphonsus, Arzachel, Hell, Hesiodus, Pitatus, Thebit Oceanus Procellarum Briggs, Cardanus, Dalton, Einstein group, Hansteen, Lavoisier group, Rep sold, Seleucus Orientale Schluter Serenitatis Posidonius Tranquillitatis Ritter, Sabine, Vitruvius

Komarov, a large crater with a cracked floor located on the lunar farside (Fig. 68), is an excellent example of the occurrence of 118

Figure 68. Mare Moscoviense 119 such craters along the margins of the circular maria. In this example,

KOmarov is situated over the second major structural ring encircling.

Mare Moscoviense. In other examples, craters with cracked floors are found within the maria close to the margins or else in the uplands im­ mediately surrounding the basins; the crater Lock is an example of the former and Vitello of the latter. Another common mode of occurrence is for a crater with a cracked floor to be appreciably tipped toward the cir­ cular mare and to be partially engulfed, by mare materials; the craters

Gassendi, Doppelmayer, Pitatus, Posidonius, and Cassini fall within this category. Craters with cracked floors are also found in similar positional relationships along the margins of the irregular maria. In Figures 69 and 70, a view of the western margin of Oceanus Procellarum, a concentra­ tion of craters with cracked floors occurs along the upland-mare contact.

Similarly, in Figure 71, the craters with cracked floors are located in the zone of tonal transition that marks the edge of Mare Smythii.

Formation of the Circular Maria

The origin of the circular mare basins has been quite controver­ sial, but the detailed studies of Mare Orientale by Hartmann (1964) and

Hartmann and Yale (1969) have favored a catastrophic asteroid impact on the lunar surface as an initiating mechanism. Of all of the circular mare basins, the Orientale structure appears to be the youngest and least filled with mare materials. Hartmann attributes this lack of filling to a decreased quantity of near-surface magmas available at the time of the impact (the impact having occurred after the period of maximum internal Figure 69. Northwestern Margin of Oceanus Procellarum 121

Figure 70. Map of Northwestern Margin of Oceanus Procellarum Derived from Figure 69

Circles are craters with cracked floors. Figure 71. Mare Smythii

Source: Orbiter II photograph 192

Scale: 1 cm equals approximately 29 km General location: Eastern lunar limb region

Surrounding Mare Smythii there are at least ten craters with cracked floors . Within these craters the cracks occur only in the lighter and more hummocky materials, while the uncracked, dark marelike materials fill the lowest portions of the crater floors and often cover the cracks. Several of the dark-filled craters may represent a complete covering of cracked floors.

North is toward the top of the photograph. Figure 71. Mare Smythii 123 heating of the moon through the decay of radioactive isotopes. Because, of the youth and low degree of mare filling, the multi-ring structure of the Orientate basin has not been obscured by subsequent events

(Fig. 72). Mackin (1969) proposes that the major ring breaks surrounding

Mare Orientate are the result of rebound fracturing which occurred im­ mediately after the dissipation of the initial radial compressive shock wave. Mackin also postulates that the separation of the rings and the development of the ring scarps were facilitated by the existence of a subsurface, nearly molten layer of rock upon which the ring blocks slid and then differentially settled. In the excavated central portion of the impact structure the molten layer was exposed and free to fill the rela­ tively small depression. The major circular fractures separating the ring blocks also reached the nearly molten layer and the sudden drop of pressure which accompanied the formation of the fractures was sufficient to allow the formation of magma. This newly formed magma then utilized the fractures in reaching the lunar surface. Magmas associated with these ring fractures created the arcuate patches of mare material (Mare

Veris and Mare Autumni) which encircle Mare Orientale.

If such a picture can be applied to the other circular mare basins in which filling has been more pronounced, it would seem that the major ring breaks would be ideal sites for the eruption of lavas. As the major bulk of a mare cooled and shrank, tension would build up in the fractured margins of the basins. Periodically, the ring fractures would open up, allowing magmas that were either present at depth or were generated by the sudden relief of pressure to extrude onto the 124

Figure 72. Mare Orientale 125 surface. These magmas would be less dense than the cooled mare mater­ ials and if released in sufficient quantity could pour out of the ring frac­ tures and inundate the low-lying mare surface. Such a process would resemble a chain reaction, for as more lava was erupted at the lunar surface, the mare surface would subside in correlation to a depletion of the underlying reservoir. Further subsidence, in turn, would open the ring fractures still wider, speeding the flow of magma to the surface.

Finally, eruption would cease due to depletion of heated materials or to a sufficient relief of tension.

Such a volcanic-tectonic process could occur on a periodic basis for a particular circular mare with each succeeding eruption taking longer to occur and then erupting less materials as the source is depleted.

In the final stages of such a process, successive eruptions might be separated by millions of years and eruptive sites would shrink from large arcuate sections of the ring fractures to small, isolated vents.

Furthermore, differentiation of magmas would be expected to occur during the long periods of tectonic inactivity, resulting in the final stages of mare volcanism being marked by small-scale, cyclic, and diverse vol- canism within the ring fracture zone.

In support of this model, workers like Hartmann (1964), Fielder

(1965), Quaide (1965), Ronca (1965), Titley (1967), and Adler and Salis­ bury (1969) have all pointed to the margins of the circular maria as regions of tension with accompanying tectonism. The circum-mare floored rilles with their graben morphology, the sets of irregular rille . systems paralleling the mare margins, and the scarps, dissected craters, and polygonal craters at the mare margins are all tectonic expressions of 126 this tension. In addition, the occurrence of sinuous rilles along the margins of the circular maria, as observed by Holcomb (1969, personal communication), also tends to support this picture, for the sinuous rilles are endogenetic features that often postdate the last stages of mare filling. Finally, during the Apollo 15 lunar expedition it was ob­ served in the walls of the Hadley Rille that mare materials in Palus Putredinis consisted of diverse, layered volcanics which strongly sug­ gest cyclic volcanism.

Formation of the Irregular Maria The irregular maria differ from their circular counterparts in that they appear to have been generated entirely by internal lunar pro­ cesses. Large blocks of cratered upland materials apparently have sub­ sided and the resultant, low-lying areas were then covered by mare materials. The amount of subsidence can often be estimated, for the tops of old, large, pre-mare craters often project above the mare filling.

The outline of an irregular mare depends upon the shape of the sunken block which, in turn, may reflect the areal extent of the underlying zone of high heat flow and preexisting regional fracture patterns. The actual process of subsidence and filling is probably identical to the filling of the circular maria with the important exception being the lack of regular ring fractures. As an irregular crustal block sinks and forms an irregular­ ly shaped basin, tension.is still found along the margins of the basin, but instead of the creation of new fractures, preexisting major faults and regional fracture systems are opened by the tension.

\ 127 On the western margin of Oceanus Procellarum subsidence has taken place along the north-northeast component of the lunar tectonic grid „ Figure 73 (in pocket) is a map of the area shown in Figure 1. At­ tention is drawn to the occurrence of craters with cracked floors along the major lineaments that parallel the mare margin. This situation is exactly as predicted by theory. Given a strong regional lineament or fracture pattern, subsidence of a crustal block occurs along preexisting fractures, the resultant tension opens up additional fractures parallel to the subsiding block, and, if magma is available at depth, the sym­ pathetic fissures will channel lavas to the surface and into large craters.

The final expression of volcanism, therefore, depends on the structure of the crater into which the lavas have been injected.

Lineaments Related to Craters with Cracked Floors

Fissures, lineaments, and major faults have been called upon as eruptive vents to supply crater floor fillings and even to initiate later- stage volcanic-tectonic processes capable of producing cracks. If a linear structure does supply magma to the lunar surface, it then would be expected that the resultant volcanics would tend to obscure this lineament.

Patches of dark marelike materials in Humboldt are found where prominent regional lineaments cross the periphery of the crater floor

(Figs. 18 and 19). These dark fillings are uncracked and show no signs of the regional lineaments on their surfaces.

The eight small cracked craters surrounding Einstein A all fall upon major north-northeast-trending lineaments (Figs. 74 and 75), but Figure 74. Crater Einstein A and Eight Associated Minor Craters with Cracked Floors Figure 75. Map of Crater Einstein A and Eight Associated Minor Craters with Cracked Floors Derived from Figure 74

Key to Map Symbols cch crater chain ic indistinct crater cp central peak cr crater rim material Ifm light-colored floor cw crater wall material dfm dark floor material sc subdued crater dm dome url unspecified rille fc fresh crater 129

cw If m c r If m \ url

rdf nr

Ifm c w Einstein A Ifm

dfm cr cp cp

c w Ifm url

cr

Ifm lew Ifm

cch

sc

Figure 75. Map of Crater Einstein A and Eight Associated Minor Craters with Cracked Floors Derived from Figure 74 130 the floor fillings and cracks within these craters (Figs. 22 and 76) do

not reflect the linear trend of the lineaments „

The rims of the craters Alphonsus (Fig. 23), Airy (Fig. 41), and Hesiodus (Figs. 49 and 50) are also cut by major rilles which either do not continue across the crater floors or else do so in a highly subdued manner. The large crater chain in Mersenius (Figs. 51 and 52), how­ ever, might represent the elongated source vent. In this last example, note that the cracks which postdate this crater chain are not parallel to this lineament.

In chapter 2 it was shown, as well as in the examples just cited, that the cracks within crater-floor materials are rarely aligned parallel to major lineaments which cross the crater rims. From this re­ lationship it may be inferred that either negligible tectonism has oc­ curred along the lineament since the emplacement of a volcanic filling within the crater or that the strength and coherence of the filling material has resisted deformation during minor post-filling tectonics. If the

second of these two alternatives has occurred, the crater filling material can be considered to be mechanically dis coup led from the original crater floor and from regional tectonics in general. Stresses capable of cracking the strong floor filling would have to affect large portions of the crater floor.

Summary

Craters with cracked floors commonly occur along major linea­ ments and in many instances it can be shown that these lineaments occur in the regional zones of tension that encircle the circular and 131

Figure 76. Craters Einstein "W" and Einstein "X" 132 irregular maria. Volcanism is anticipated along these lineaments on a theoretical basis in which topographic depressions, especially craters, would be the more probable sites of eruption. The postdating of the lineaments by the crater-floor fillings is consistent with the model that these filling materials were derived from the lineaments. CHAPTER 7

TERRESTRIAL ANALOGUES

No specific terrestrial analogue exists for the wide, shallow cracks found on the floors of lunar craters. The fractures found in the frozen crust of the lava lake in Halemaumau in Hawaii display similar fracture patterns but are not comparable to the lunar cracks because of their small size„ Quite possibly, wide cracklike features may not be produced on earth, even if similar tectonic and volcanic conditions were present. A brittle plate, similar in size and strength to the fillings in lunar craters, might deform differently if arched in a much stronger gravitational field, because a greater number of major fractures with less displacement along each fracture would be expected due to a higher lithostatic pressure. Erosion would also quickly modify any long slopes composed of a poorly consolidated pyroclastic debris. Considering dif­ ferences in the geologic environment, terrestrial features corresponding to the lunar cracks might be wide fracture zones, grabens, dikes, or fault-controlled drainage patterns.

Due to the absence of wide, cracklike features, terrestrial analogues to the lunar craters with cracked floors must be distinguished on the basis of secondary features. Three features that terrestrial ana­ logues may be expected to display are: 1. A depression diameter similar in magnitude to the diameter of

the lunar craters with cracked floors,

133 134 2. A complex history of successive volcanic floor fillings with emphasis on pyroclastic eruptions, and

3. Central subradial and peripheral concentric fracture systems that may be associated with updoming or subsidence.

The only terrestrial features, that fulfill all these conditions are the resurgent calderas.

Resurgent Calderas

Terminology and General Characteristics According to Smith and Bailey (1968, p. 617-617),

A resurgent cauldron /circular cauldrons are calderas/ is defined as a cauldron within which the. cauldron block /roof of magma chamber/, after initial subsidence, has been uplifted, usually in the form of a structural dome. The structural dome, broken radially or concentrically, or both, commonly shows secondary collapse features, or the effects of differential move­ ments of major segments, related to the uplift, resurgent caul­ drons commonly, but not necessarily, exhibit synresurgence and post resurgence volcanism along the cauldron ring frac­ tures or along the fractures within the structural dome or both.

In addition, resurgent calderas are characterized by massive outpour­ ings of.ash (often in tens to hundreds of cubic kilometers) which are products of a silicic, viscous volcanism.

The above description is for a purely volcanically generated structure, but if the term "crater-floor filling" is substituted for

"cauldron block," the post-subsidence resurgence and associated vol­ canism described above would also apply to the cracked floors of lunar craters which may have originated as impact structures instead of volcano-tectonic collapse depressions. A discussion of similarities between the brittle floor fillings of lunar craters and the collapsed 135 cauldron blocks of terrestrial calderas will be presented later in this chapter.

Resurgent calderas are the largest circular volcanic depressions found on earth, and they are of the same order of magnitude as the lunar craters with cracked floors (Table 5). Terrestrial resurgent calderas are somewhat smaller than their lunar counterparts, but this may be expected for two reasons. First, assuming plates of equal thickness and vertical­ ly directed forces of equal intensity on both the moon and earth, plates six times larger in area could be uplifted on the moon due to the lower gravitational field. This increase in plate area would correspond to an increase in plate diameter by a factor of approximately 2.5, which, when applied to the 15 to 20 kilometer diameters of terrestrial resurgent calderas, would predict the 37 to 50 kilometer crater diameters observed on the moon. Second, the dimensions of the subsided cauldron block in terrestrial calderas is limited by the size of the underlying magma cham­ ber, whereas on the moon the area of the crater-floor filling might be determined either by the dimensions of a volcanic collapse depression or a much larger meteoritic impact. However, the minimum size of 13 km for terrestrial resurgent calderas, as observed by Smith and Bailey

(1968), is larger than the 9-km minimum diameter for lunar craters with cracked floors. If lunar craters with cracked floors and terrestrial resurgent calderas are genetically related to each other, then the nature of internal structures and geologic histories of lunar craters with cracked floors may be inferred from the eroded remains of terrestrial calderas. Brief des­ criptions of three terrestrial resurgent calderas, the Valles, Timber 136 Mountain, and Turkey Creek calderas follow, with a comparison between the resurgent calderas and lunar craters with cracked floors.

Table 5. Examples of Terrestrial Resurgent Cauldrons and Calderas

Caldera Location Dimensions (km)

Toba (volcano-tectonic) a depression Northern Sumatra 96 X 29 a San Juan Colorado 48 X 24

Timber Mountain Nevada a 32 X 29 a Long Valley California 29 X 16

Valles New Mexico a 24 X 19

Creede Colorado a 22

Turkey Creek Arizona b 21 a Silverton Colorado 18 X 13 a Lake City Colorado 14 X 10

a . From Smith and Bailey (1968). b. From Marjaniemi (1969).

Valles Caldera

The Valles caldera (Fig. 77) is located in the Jemez Mountains of northern New Mexico. During mid-Pleistocene times, two major, catastrophic eruptions of rhyolitic ash occurred. The first eruption, which correlates with the first cycle Bandelier Tuff, produced the Toledo caldera. Some 300,000 years later, the second eruption created the Valles caldera which partially overlaps the Toledo structure. The 137

<5

Tv

■'W. As v'"

Contact Alluvium Early rhyolite Topographic rim of caldera

Caldera fill Pre-Tcrtiary rocks 4 m i l e s

Figure 77. Geologic Map of the Valles Caldera, New M exico.- from Smith, Bailey, and Ross, 1961, p. 147. 138 eruption of the second cycle BandeHer Tuff (nearly 200 km3), associated with the evacuation of the underlying magma chamber, caused the roof to subside as a nearly intact, block (24 by 19 km) along a zone of ring fractures, forming the Valles caldera. During subsidence a large portion of the erupting rhyolitic ash settled within the depression to create a thick welded tuff cover over the cauldron block (Smith, Bailey, and Ross,

1961, and Smith and Bailey, 1968). Post-subsidence tectonism then arched the floor some 900 meters, breaking it up into a mosaic of blocks

(some of which dip radially as much as 25 degrees). Concurrent and subsequent to resurgence, volcanism occurred with a centrally located graben and in the 3 to 5 kilometer wide ring-fracture zone at the base of the dome. A more detailed outline of the structural and volcanic develop­ ment of the Valles caldera appears in Table 6. Post-subsidence volcanism in the Valles caldera is marked by three main stages:

1 „ Early rhyolitic eruptions of variable character that contributed

little to the floor filling,

2. Middle rhyolitic eruptions within the central graben that oc­

curred during doming, and

3. Post-doming, late rhyolitic eruptions in the moat area that

created ten rhyolitic domes, ranging between 1 and 3 km in diameter.

Timber Mountain Caldera

The Timber Mountain caldera in Nevada (Fig. 78), as described by Smith and Bailey (1968, p. 619), is a mid-Pliocene volcano-tectonic 139 Table 6. Structural and Volcanic Development of the Valles Caldera , ' New Mexico

Stage Structural Events Volcanic Events Duration

I. Broad regional doming over Eruptions due to less than magma chamber with ac­ leakage along 4 x 105 years companying formation of ring or radial ring and radial fractures fractures

II. Collapse-subsidence of Main eruption of less than 16-km-diameter roof Bandelier Tuff 10 years block along ring frac­ (200 km3) into ture s depression and surrounding region

III. Avalanches and debris flows Pre-resurgence less than onto floor from caldera volcanism of 105 years walls of early rhyo- lites; minor lava and pyroclastic eruptions mixed with lake-bed deposits

IV. Resurgent doming of floor Eruption of middle less than with development of rhyolite lavas IQS years central graben and ring from central fractures graben and north­ west portion of ring fracture zone

V. Re-opening of ring frac­ Eruption of late approximately tures after cessation rhyolite s in the 8 x 105 years of floor doming form of cones, . domes, and flows . in the ring frac­ ture zone

VI. . Quiescence Terminal fumarolic greater than hot spring ac­ IQS years tivity in moat and central graben 140 iZ

RAINIER / MESA

rNortheastern

■ i r TIMBER MOUNTAIN to / r

V v 't

EXPLANATION

Post-Timber Mountain units and post collapse lavas (includes moat ESS rhyolites)

Timber Mountain Tuff (includes Rainier Mesa Member)

Paintbrush Tuff and precollapse lavas

6 MILES Pre-Paintbrush units

Figure 78. Geologic Map of the Timber Mountain Caldera. — from Christiansen et al. , 1965 # p. 45. 141 collapse structure, slightly elliptical in shape (32 by 29 km) that „ „ . formed by subsidence during and after the eruption of ap­ proximately 250 to 500 cubic miles (1000 to 2000 km3) of rhyolitic ash flows. Doming of the caldera floor, accompanied by minor rhyolitic intrusions and effusions from fractures in the dome, followed. Finally, rhyolites erupted from the ring- fracture zone in the moat." The Timber Mountain dome shows extensive radial and concentric distension faulting with well defined radial and apical grabens.

Carr (1964) and Christiansen et al. (1965). add to this picture several important details reminiscent of features observed in the lunar craters with cracked floors:

1. The updomed portion of the caldera floor is a composite of

several closely related ahs flow tuffs that were erupted during the collapse of the magma chamber and in the initial stages of

resurgence (welded pyroclastic filling).

2. One of the earliest structures to form during updoming is a 1.5-

km-wide zone of fracturing surrounding the base of the dome;

covering this fracture zone are silicic flows, basalts, tuffs,

tuffaceous sandstones and conglomerates, and colluvium

(concentric fracturing and diverse moat volcanics).

3. During updoming, rhyolite erupted from the central graben, while • granite porphyry intrusions were emplaced parallel to the ring fractures (fissure-controlled volcanism).

4. Potassium-argon dating of the ash flow tuffs in the moat show

that the Timber Mountain caldera was still volcanically active

3.5 million years after the initial subsidence (long volcanic

history)„ 142 5. Fault intersections between ring fractures and regional (Basin

and Range) faults were favored sites for volcanic eruption

(structural control of volcanism by regional faults).

6 . Few regionally controlled faults occur within the Timber Moun­

tain caldera (structural discoupling between caldera filling and

regional tectonics).

Turkey Creek Caldera

The Turkey Creek caldera is a highly eroded collapse structure, some 22 km in diameter, that is located in the Chiricahua Mountains of southeast Arizona. Differential erosion has removed portions of the caldera rim as much as 900 meters beneath the more resistant rhyolitic moat deposits, creating an erosional inversion of topography. Marjaniemi

(1969) was only able to identify the structure on the basis of differences in rock type, the presence of ring faults, and stratigraphic discontinu­ itie s, such as are observed in the Rhyolite Canyon Tuff, which is 900 meters thick within the caldera but only 240 meters thick outside the caldera.

Initial collapse occurred approximately 24.9 milliong years ago, when an estimated 400 km3 of Rhyolite Canyon Tuff were erupted. Sub­ sequent updoming of the floor was caused by the emplacement of a fine­

grained quartz monzonite which was apparently injected between large fault blocks, often greater than 1. 5 km across. Dikes , 3 to 30 meters wide , were also emplaced as the thick tuff filling in the caldera arched. Subsequent to doming, moat rhyolites, consisting of several 143 tuffs and flows were unconformably deposited over the Rhyolite Canyon Tuff in the moat region surrounding the domed floor.

Comparison Between Terrestrial Resurgent Calderas and Lunar Craters with Cracked Floors

As summarized in Table 7, many of the lunar craters with cracked floors display features that are characteristic of resurgent cal­

deras. The similarity is so great in many examples that one is forced to admit similar modes of development for many of the lunar and terres­ trial features, but the presence of lake sediments in the moat regions of

several terrestrial calderas (Smith and Bailey, 1968) indicate that several

major differences may exist between the terrestrial and lunar structures.

For example, the extensive hydrothermal alteration and associated hot- spring activity observed in the terrestrial resurgent calderas is a product of residual volcanic heat and abundant meteoric water percolating through the ring fracture zone. Unless the early lunar environment contained far more water than is generally postulated, similar zones of alteration would not be expected on the cracked floors of lunar craters. Erosion of terrestrial resurgent calderas during doming would also produce struc­ tural differences. Volcanism and fracturing at the apex of a domed ter­ restrial structure would be expected to be more pronounced due to erosional thinning of the cauldron block overlying the magma chamber. Thick layers of sediment derived from the dome would also be expected to accumulate in the moat region.

A large number of lunar craters with concave, cracked floors

show no signs of ever having been domed and are therefore not compa­ rable to the terrestrial resurgent calderas. However, Smith and Bailey 144

Table 7. Summary of Features Common to Terrestrial Resurgent Calderas and Lunar Craters with Cracked Floors

Features Observed in Terrestrial Lunar Resurgent Calderas Analogies

Overlapping of resurgent calders, such as observed Repsold in the Toledo-Valles complex Resold G

Thick compos it of welded tuffs as the primary, Alphonsus pre-resurgent caldera filling Gassendi

Domed caldera floor, commonly with small central Lavoisier F collapse structures, such as apical grabens Petavius Hansteen

Subradial fracture pattern on updomed portion of Humboldt caldera floor Einstein "V" A wide peripheral zone of concentric fractures Lavoisier group (ring fracture zone) Pitatus

Eruption of volcanic material from central graben Petavius

Post-resurgence, diverse volcanism that obscures Humboldt fractures in the moat region Petavius

Breaking of resurgent dome into a mosaic of large Gassendi blocks Lavoisier H

Cones and rhyolitic domes, 1-3 km in diameter, Alphonsus structurally located over fractures Hesiodus

Eruption of volcanic material from the intersection Humboldt of regional faults and the ring fracture zone

Few regional faults crossing caldera fillings Lavoisier group 145

(1968) describe a rare variety' of terrestrial calderathe Glen Coe type, in which the cauldron block slowly subsides as lavas are erupted from the ring fracture zone. If, in this particular type of caldera formation, the central portion of the cauldron block subsides faster than its periph­ ery, flexure and accompanying fracturing would occur along the margins of the cauldron block, creating a belt of concentric fractures. Lunar craters, like Pitatus, Hesiodus, and Lavoisier with their concave floors, might be lunar equivalents of Glen Coe-type calderas.

Cauldron Block vs Crater Filling

Some authors , such as Smith and Bailey (1968), describe resur­ gence in terrestrial calderas as the updoming of a nearly intact cauldron block, but in most examples no trace of the cauldron block is observed.

Instead, the domed portion of the caldera floor commonly consists of a thick welded tuff sheet that was emplaced during the initial subsidence of the caldera. Marjaniemi (1969) in his study of the Turkey Creek cal­ dera suggests that only the welded tuff filling and not the cauldron block may be involved in doming. This view is supported by the direct contact between the Rhyolite Canyon Tuff and the quartz monzonite intrusion which was responsible for updoming in the Turkey Creek caldera. There­ fore, the welded tuff fillings in terrestrial resurgent calderas may pro­ vide a closer analogy to the brittle fillings of lunar craters than the buried cauldron blocks. 146 Physical Characteristics of Magmas Associated with Lunar Craters with Cracked Floors

The volcanism of terrestrial resurgent calderas is quite silicic and viscous, starting with the massive outpourings of tuffs erupted dur­ ing the collapse of the cauldron block and continuing through the forma­ tion of post-doming rhyolitic domes in the moat area. Since lunar craters with cracked floors display many of the features characteristic of these resurgent calderas, it is reasonable to predict that similar volcanic conditions may have existed on the moon.

The features associated with Damoiseau (Fig. 79) are best explained by a collapse origin in which large quantities of ash were erupted. The light-colored, subdued topography of the uplands sur­ rounding Damoiseau could be readily explained as a thick tuff sheet originating within Damoiseau. Probably the low raised rim is also a volcanic features. It would be difficult to explain this low rim as the subdued rim of an impact crater, since the rim of the older, larger crater in which Damoiseau is located is much thicker and has not lost its meteoritic impact appearance. Finally, the irregular plates within

Damoiseau that are separated by cracks suggest the broken remains of a collapsed cauldron block. Quite possibly, in its present state,

Damoiseau may resemble the Timber Mountain caldera shortly after it collapsed. With just slight changes in the above picture, Damoiseau may also be viewed as the updomed and collapsed floor of the older, larger crater, once again linking the volcanic and impact structures.

Damoiseau displays many of the features commonly associated with silicic terrestrial calderas, but these features may be the result 147

5 #

mm

i S t o a ,

Figure 79 . An Oblique View of Damoiseau 148 of magma viscosity rather than silica content. According to Smith and

Bailey (1968, p. 650), the viscosity of a magma should affect the mode of volcahism within a terrestrial caldera, in that

. . . high-viscosity magma would tend to distribute magma pressure over the base of the subsided cauldron block and thereby promote uplift and doming. Low viscosity magma, in contrast, would tend to dissipate magma pressure more rapidly by flowage through fissures in or adjacent to the cauldron block with consequent surface volcanism. On earth, magma viscosities commonly correlate with silica content, but on the moon the volatile content of a magma may be more crucial. A lunar magma low in volatiles might be quite viscous, but the near-vacuum environment at the lunar surface might still favor de­ gassing with possible pyroclastic eruptions in spite of.the low gas con­ tent of the magma. Consequently, only the viscosity of a lunar magma may be inferred from the landforms it creates. Whether or not the mag­ mas responsible for cracking within lunar craters were either silicic or low in volatiles can only be determined when materials from these craters are sampled.

Martian Analogues to the Lunar Craters with Cracked Floors The prerequisites for terrestrial analogues to the lunar craters with cracked floors were quite simple. A major magmatic source (prob­ ably viscous) had to develop beneath a large depression that already contained a brittle, coherent filling. No mention is made in these pre­ requisites as to the origin of the depression or to the nature of material that fills it. The depression may have developed either as a volcano- tectonic subsidence or an asteroid impact structure. 149 Considering the relatively high abundance of large circular depressions and feature of undisputable volcanic origin on Mars, as recorded by the Mariner DC probe, it is no surprise that the newest photographs of the Martian surface also show features that strongly resemble the lunar craters with cracked floors. Photographic resolution, in general, is too low to distinguish individual cracks, but on the floors of several Martian craters there are distinct changes in albedo that are comparable to the volcanic moat deposits in both the updomed floors of lunar craters and terrestrial calderas. If the high-resolution photographs of these particular Martian craters reveal that the crater floors are frac­ tured like the lunar craters and terrestrial calderas, it will be difficult to deny that the volcanic and tectonic modification of caldera and im­ pact crater floors is not just a coincidental event but rather a relatively common geologic process. CHAPTER 8

A GENERALIZED MODEL FOR THE FORMATION OF , LUNAR CRATERS WITH CRACKED FLOORS

As indicated throughout this thesis, both meteorite impacts and various forms of volcanism have played major roles in creating the pres­ ent morphology of the lunar surface. In this light, the lunar craters with cracked floors are important features for they appear to be a combination of the.results of both impact and volcanic processes . The lunar craters with cracked floors also form a continuous series with the large circular maria, and, hence, a study of these smaller but far more numerous, cra­ ters with cracked floors may provide a basis for interpreting the evolu­ tion of the lunar maria. The following discussion presents a generalized model for the formation of lunar craters with cracked floors.

Crater Formation

The initial stage in the development of a lunar crater with a cracked floor is the formation of the depression either by meteoritic im­ pact or by the collapse of the cauldron block overlying a magma chamber.

In either event, a circular depression with slumped wall blocks will re­ sult along with a crater filling of fine debris and a mantling of this debris over the rim of the crater. Although the dimensions of depressions created by these two processes may be quite similar, differences in the mode of origin can severely affect the manner in which the depressions are later modified.

150 151 An impact origin is more probable for the large lunar craters be­ cause this mechanism is not limited by energy considerations such as those that must be applied to the formation of calderas. When an impact­ ing me teoritic body strikes the surface of a planet, it is often traveling faster than the seismic velocity of the material that it encounters. As a result, solid matter comprising the planetary crust is suddenly shoved aside and radially compressed under tremendous pressures often exceed­ ing 1,000 kilobars. When the meteorite body finally comes to rest, a large portion of its total kinetic energy has been converted to the com­ pression of the crystalline structures of minerals in the crustal rocks.

This condition exists but momentarily and is immediately followed by a tremendous implosion as the compressed crystalline structures seek to relieve their strained configurations. During the implosion, the com­ pressed materials expand back toward the point of impact, expanding so fast that rocks are disrupted by release fracturing. These fragmental materials, traveling at high velocities, meet at the center of the impact site and are ejected upward and outward by their explosive force, ex­ cavating the centraImost portion of the impact site. Materials further from the center of the impact are compressed less and hence develop fewer release fractures during implosion. The removal of lateral support as materials are ejected from the center of the crater, when coupled with the production of concentrically oriented release fractures, promotes the extensive slumping of wall blocks in the larger impact structures.

If the impact has been in relatively cold crustal rocks, the crater floor should consist of a layer of fine, fall-back debris that grades downward into coarser and coarser breccias, as shown in Figure 152 80. If the crustal materials are inhomogerieous and contain resistant structures, such as igneous intrusions, then the impact crater floor might resemble the floor of Tycho (Fig. 42), and could large hills, consisting of these resistant structures, surrounded by the breccia filling. In contrast, if the impact has been of sufficient magnitude to expose submolten crustal rocks, which might exist some .10 to 30 km be­ neath the planetary surface at the time of impact, the resulting crater floor would be different. The sudden relief of lithostatic pressure, fol­ lowing excavation, would allow partial melting of these submolten rocks which, in turn, could create a welded breccia on the floor of the crater.\

Finally, the breccia filling within an impact crater and the ejecta blanket surrounding it would be characterized by rock fragments mixed with a small quantity of material that was fused during the impact of the meteorite.

The formation of a large caldera in which the roof of the magma . chamber has subsided along ring fractures while ignimbrites were being erupted would produce features distinctly different from those which characterize meteoritic impacts. For example, the ignimbritic material would be erupted in a molten state, and, although it would be expected to quickly cool, the resulting mantle of this material surrounding the caldera would contain few rock fragments, since the extensive release fracturing present in impact implosions does not operate here. The materials comprising this ignimbrite sheet would perhaps resemble the welded tuffs on earth in which the bottommost zones of the tuff sheet are compressed and welded, while the uppermost zones are only semi­ consolidated. Depending on the history of the magma chamber from which ' DEPRESSION

o' b'.-o. S tiS lo

Figure 80. Idealized Structure of Large Impact Crater Rim deposits have been left off. 154 these materials were erupted, a porphyritic texture of these pyroclastic materials might also be anticipated. Structurally, the floor of a caldera would also differ from an impact crater for it should contain a fractured cauldron block capped by a thick covering of ignimbrites. Less slumping of wall blocks would also be expected since release fracturing is not an important process in cal­ dera formation. Ring fractures along which a cauldron block subsides i - might resemble the concentric, multi-ring structures of large impact craters, but fragments of the cauldron block would tend to be tipped toward the center of the crater in contrast to the slight backward rota­ tion of wall slump blocks in impact craters.

In spite of these significant petrologic and structural differ­ ences between large impact craters and large calderas, depressions re­ sulting from either process would have a poorly consolidated material on the crater floor. An impact crater would contain a breccia blanket and the uppermost zone of the pyroclastics within a caldera would be unwelded.

Crater Fillings

The shallow profiles of most lunar craters with cracked floors suggest that these craters have been filled with materials that postdate crater formation. From the presence of cracks within the crater floor materials, it may be inferred that these materials are brittle and lateral­ ly homogeneous. In several examples, these materials may represent' the original welded tuff cover over a collapsed cauldron block or else a primary welded breccia that was emplaced at the time of impact, but, in 155 general, a secondary volcanic filling must be postulated for the crater-

floor materials. This secondary filling may consist of the emplacement of fluid lavas, resulting in a crater floor like Tsiolkovsky's (Fig. 43) , or else ignimbrites, as observed in Alphonsus (Fig. 44). The secondary

filling may even be multiple and complex, representing several, sepa­ rate periods of volcanism. The source for these secondary fillings is relatively easy to

identify in the calderas, since they have already formed over a mag- matic vent, but for the meteoritic impacts which are randomly distributed over a plant, volcanic fillings would be expected only where the impact coincides with a volcanic region, such as along the margin of a mare.

A meteoritic impact may expose an igneous intrusion and thus immedi­ ately trigger post-impact volcanism or else the conditions favoring vol­ canism may occur long after the formation of the crater; in either event, the impact crater will be filled by post-imp act volcanic s .

Figure 81 presents just one of the many possible modes in which a secondary volcanic filling may be emplaced within an impact caldera.

If. large quantities of the lava are erupted in a short period of time, the floor of the crater could become a lava lake over which a crust would

slowly form.. This figure also illustrates how the structure underlying a lunar crater could affect the mode of eruption; upward-welling magmas would tend to travel along the concentrically oriented, release fractures to erupt along the periphery of the crater floor. The magmas might also

serve as a lubricant in these fractures and thus cause a renewal of wall- block slumping and, consequently, an increase in crater diameter. FILLING

Figure 81. Development of a Chilled Crust over a Post-impact Lava Crater Filling 157

In short, there are many possible ways in which a brittle vol­ canic filling might be emplaced within either a caldera or impact crater.

Development of the Cracks

The cracks observed on the floors of lunar craters appear to be fractures created during the uplift or collapse of a brittle filling within a crater. The extent and pattern of the cracks depend not only on the di­ mensions and general character of the materials comprising the fillings but also on the degree of uplift or collapse to which the filling is sub­ jected. A thick, rigid filling might show only minor cracking in spite of major uplift, whereas a thin, less rigid filling would quickly respond to even small tectonic adjustments beneath the floor of the crater.

Considering the high concentration of lunar craters with cracked floors along the margins of the major maria and the presence of crater- floor materials that fill and therefore postdate the cracks within these craters, it can be inferred that both uplift and.collapse of the crater floors are caused by volcanic intrusions associated with the development of the adjacent maria. In Figure 82, one possible cause for doming and cracking is illustrated in which the youngest volcanic material forms a laccolithic intrusion beneath an older floor filling. If, perhaps, the youngest magma had been less viscous in this example, doming might not have occurred and more lavas might have been erupted from the pe­ riphery of the crater floor. In still another example, the central portion of the crater floor might collapse following subsurface drainage of mag­ mas, shrinkage of the underlying magma upon cooling, or minor eruptions of ignimbrites from the crest and flanks of the dome. A large variety of UPDOMING OF FLOOR

Figure 82. Updoming of the Crater Floor by Laccolithic Intrusion and the Emplace­ ment of Moat Volcanics 159 possible volcanic modifications of the crater floor are possible, but doming would be more commonly associated with viscous magmas and subsidence with the more fluid ones „ Cracking within the largest lunar craters might be more of an iso static adjustment process in which a slowly heating planetary crust gradually loses its high viscosity, allowing old impact depressions to level out. The initial stages in such an adjustment would include a low doming of the crater floor with the possible production of cracks and ex­ tensive slumping of the crater walls .

Implications

The large, post-basin, pre-mare craters with cracked floors that are found along the upland-mare borders may provide the best in­ sight into the development of the lunar maria. The diversity of crater- floor materials of various albedos and morphologies tend to indicate that the volcanic materials erupting from the periphery of a mare have varied through time. This variation suggests magmatic differentiation and may also imply several volcanic stages in the formation of the maria in which the dark fillings are only the most recent stage. •

Geologic sampling of the floor and wall materials within any one of these lunar craters with cracked floors may provide specific dates and the nature of the materials that were being erupted into the maria.

Sampling within the crater Humboldt (Fig. 18) could provide the age of the most recent, marelike eruptions and at least five ages of pre-mare events. Samples from the crater wall should provide the age of the lunar crust, the rough crater-floor materials in the northern portion of the 160 crater floor might yield the age of the impact that formed Humboldt, and the cracked, light-colored level materials in the central portion of the crater floor would represent the major post-impact volcanic filling within the crater. Samples from the central peaks, the double-ringed crater, and the dark moat fillings might yield the dates of minor volcanic events.

Petrologic examination of these samples from Humboldt would reveal the manner in which these materials were erupted, and chemical studies would indicate the trend of pre-mare magmatic differentiation.

It is reasonable to expect structures resembling the lunar craters with cracked floors on any planetary body in which large natural depressions, either calderas or meteoritic impact craters, overlie major magmatic vents. Viscous magmas, perhaps associated with pyroclastic eruptions, would produce the closest analogies to the lunar craters with cracked floors, since less viscous lavas would tend quickly to bury the slowly subsiding crater floor. APPENDIX I

LUNAR CRATERS WITH CRACKED FLOORS

Names included within quotation marks are unofficial and have been assigned strictly for convenience . Latitude and longitude values bearing a * have been estimated.

All other latitude and longitude values have been obtained from "The

System of Lunar Craters" (Arthur et ,al„, 1963, 1964, 1965, and 1966) .

161 162

Diameter Crater (km) Latitude Longitude

Airy 37.1 1 8 .IN . 5.7 E. Alphonsus 114 13.5 S. 2.7 W. Arzachel 97.5 18.2 S. 1.9 W. Atlas 88.1 46.6 N. 44.4 E. Balboa 65.8 19.1 N. 83.0 W. Balboa "W" 36.5 *23 N. *84.5 W. Balboa "X" 32.7 *22.5 N. *85.5 W. Balboa "Y" 32.7 *22 N. *86 W. Blancanus C 38.6 66.5 S. 28.0 W. Boscovich 43.2 9.8 N. 11.1 E.

Briggs 37.3 26.4 N. 68.8 W. Byrgius D 27.4 24.1 S. 67.3 W. Campanus 43.4 28.0 S. 27.9 W. Cardanus 49.4 12.1 N. 72.4 W. Cassini 52.1 40.2 N. 4.5 E.

Cleomedes 121.8 27.7 N. 55.5 E. Colombo M 15.7 14.6 S. 47.8 E. Colombo "X" 9.3 *17 S. *47.7 E. Cruger E 15.2 17.5 S. 65.1 W. Dalton 63.8 17.1 N. 84.3 W. Damoiseau 36.5 4.9 S. 61.1 W. Daniell 26.9 35 . 3 N . 31.1 E. Davy 34.7 11.8 S. 8.1 W. Doppelmayer 68.0 28.5 S. 41.4 W. Eimmart C 23.9 22.3 N. 61.1 E.

Einstein A 50.2 16.8 N. 87.0 W. Einstein "U" 11.4 . *17.0 N. *88.5 W. Einstein "V" 21.1 *18.0 N. *88.5 W. Einstein "W" 9.7 *15 N. *89 W. Einstein "X" 12.3 *15 N. *88.8 W.

Einstein "Y" 16.7 *14.5 N. *86.3 W. Einstein "Z" 9.1 *15 N. *86 W. Fabricus 79.2 42.8 S. 41.9 E. Fontenelle 38.0 63.3 N. 18.8 W. Fracatorius 123.6 21.3. S. 33.1 E. 163

Diameter Crater (km) Latitude Longitude

Franklin 55.7 38.8 N. 47.6 E. Galilaei B 14.2 11.4 N. 67.6 W Gassendi 110.4 17.5 S. 39.8 W Gauss 173.5 35.9 N. 79.1 E. Goclenius 59.2 10.0 S. 45.0 E.

Hansteen 46.0 11.5 S. 51.9 W Hedin "X" 21.6 * 4.5 N. *82 W Hedin "W" 19.7 * 6.5 N. *82 W Hell 33.6 32.3 S. 7.8 W Hesiodus 42.5 29.4 S. 16.3 W

Hevelius 112 2.2 N. 67.5 W Humboldt 201.8 27.2 S. 80.8 E. La Condamine 37.3 53.4 N. 28.1 W Langrange W 50.6 32.8 S. 63.8 W Lavoisier 71.0 38.1 N. 80.4 W

Lavoisier A 26.4 . 36.9 N. 72.7 W Lavoisier B 24.2 39.8 N. 79 .7 W Lavoisier C 32.5 35.7 N. 76.4 W Lavoisier D 61.9 41.0 N. 77.5 W Lavoisier E 50.2 40.8 N. 79.6 W

Lavoisier F 33.2 36.9 N. 80.1 W Lavoisier H 30.2 38.0 N. 78.2 W Lavoisier "Q" 11.2 *38.8 N. *79.5 W Lavoisier " MM" 17.1 *36.2 N. *76.2 W Lick 31.4 12.3 N. 52.6 E.

" Mare Smythii B" 31.1 * 2.5 N. *84.5 E. "Mare Smythii C" . 34.2 * 1.0 N. *85.0 E. " Mare Smythii D" 38.7 * 1.0 S. *84.5 E. "Mare Smythii E" 39 * 2.5 S. *86.5 E. "Mare Smythii F" . 36 * 4.0 S. *87.5 E.

"Mare Smythii M" 70.5 * 6.5 S. *84 E. " Mare Smythii N" 45 * 6.0 S. *85.5 E. " Mare Smythii R" 30.5 * 8.0 S. *83 E. Maury A 21.0 36.0 N. 41.9 E. Mersenius 82.3 21.5 S. 49.2 W 164

Diameter Crater (km) Latitude Longitude

M essala 114 39.1 N. 59.8 E. Petavlus. 183 25.3 S. 60.4 E'. Pitatus 100.5 29.8 S. 13.5 W. Posidonius 97.1 31.8 N. 29.9 E. Repsold 105 51.2 N. 77.4 W.

Repsold G 44.0 50.4 N. 79 .4 W. Rocca L 18.4 14.1 S. 72.7 W. Riemann "X" 50.1 *37 N. *87 W. "RiphaeuS X" 43.3 * 6.5 S. *27.5 W. Ritter . 29.3 2.0 N. 19.2 E.

Sabine 29.5 1.3 N. 20.0 E. Schluter 93.9 6.0 S. 83.5 W. Schluter "X" 12.9 * 1 S. *90 W. Seleucus 44.7 21.1 N. 66.5 W. c n Struve F CD 22.5 N. 73.6 W.

Struve G . 15.3 23.9 N. 74.0 W. Stuve L 14.6 20.6 N. 76.0 W. Struve M 13.5 23.3 N. 75.2 W. Tauruntis 57.2 5.6 N. 46.5 E. Tannerus 29.1 56.3 S. 21.9 E. Thebit 56.2 22.0 S. 4.0 W. U. Beigh "X" 47.8 *27.5 N. *87 W. Vasco da Gama 98.5 13.9 N. 83.2 W. Vasco da Gama 52.0 9.9 N. 83.1 W. Vasco da Gama "V" 18.9 * 9.5 N. *79 . W. Vasco da Gama "W" 48.2 Vasco da Gama "W" 48.2 *12 N. *85' W. Vasco da Gama "X" 11.4 *12 N. *85 W. Vasco da Gama " Y" 29.5 *14 N. *84.5 W. Vitello 42.2 30.4 S. 37.5 W. Vitruvius 28.4 17.6 N. 31.2 E.

Weiss E 16.9 31.1 S. 19.2 W. Wurzelbauer 80.0 33.9 S. 15.9 W. unnamed 41.3 *17.5 S. *89.0 W. unnamed 14.0 *18.5 S. *86 W. APPENDIX II

WIDTH MEASUREMENTS OF CRACKS

The average maximum reading was obtained by taking five readings along the same largest crack and then averaging these values. The average width was obtained by taking ten measurements of the full range of crack widths within a particular crater and then averaging these values.

165 166

Average Maximum Maximum Average Minimum Width Width Width Width Crater (km) (km) (km) (km)

Alphonsus 1.24 1.23 0.82: 0.39 Arzachel 1.28 1.22 0.86 0.39 Atlas. 2.47 2.34 1.38 0.52 Balboa 3.08 3.08 1.71 0.66 Balboa "W" 1.95 1.41 0.98 0.39

Balboa "X" 1.58 1.58 0.91 0.56 Balboa "Y" 1.12 1.10 0.80 0.41 Blancanus C 0.79 0.71 0.53 0.28 Boscovich 1.45 0.99 0.74 0.28 Briggs 1.38 1.20 0.72 0.28

Byrgius D 1.21 0.98 0.70 0.39 Cardanus 1.10 0.71 0.63 0.14 Cassini 2.48 2.23 1.36 0.62 CrugerE 0.96 0.94 0.77 0 .54 Dalton 2.38 2.07 1.61 1.04

Daniell 1.59 1.52 1.01 0.53 Doppelmayer 1.65 1.64 0.95 0.42 Einstein A 1.10 1.05 0.43 0.20 Gauss 2.48 1.80 1.29 0.47 Hansteen 1.52 1.50 0.77 0.35

Hedin "X" 1.68 1.38 0.92 0,60 La Condamine 2.14 1.52 0.89 0.29 Lagrange W 0.93 0 .93 0.53 0.25 Lavoisier 1.72 1.65 0.99 0.55 Lavoisier A 1.45 1.40 0.80 0.25

Lavoisier B 0.76 0.76 0.52 0.29 Lavoisier D 2.09 1.95 0.93 0.42 Lavoisier E 1.47 1.18 0.92 0.29 Lavoisier H 1.46 1.06 0.87 0.48 Lavoisier Q 1.06 1.06 0.67 0.35

Mersenius 1.42 1.38 0.85 0.33 M essala 2.82 2.26 1.36 0.57 Petavius 2.84 2.84 1.29 0.42 Pitatus 2.62 2.22 1.21 0.39 Posidonius 1.24 1.14 0.85 0.33 167

Average Maximum . Maximum Average Minimum Width Width Width Width Crater (km) (km) (km) (km)

Repsold 3.21 2.29 1.17 0.52 Repsold G 2.05 2.04 1.00 0.46 Riemann "X" , 1.00 0.91 0.77 0.31 Struve G 1.16 0.96 0.68 0.40 Tannerus 1.12 0.82 0.70 0.28

The bit 2.42 2.42 0.91 0 .35 Vasco da Gama 2.00 1.80 1.23 0.54 Vasco da Gama R 1.95 1.72 1.02 0.30 Vasco da Gama "V" 0.92 0.83 0.66 0.37 Vasco da Gama "W" 1.54 1.47 1.14 0.54

Vitello 2.23 2.23 0.92 0.35 Wurzelbauer 2.14 2.14 1.10 0.72 Average values: 1.71 1.51 0.72 0.40 APPENDIX III

DEGREE OF CRACKING

The Cracked Ratio is simply the ratio between the total length of cracks within a crater and the total area contained within the rim crest of that crater. It is a measure of the degree of cracking, namely, crack length per unit area.

168 169

Total Length of Cracks Cracked Crater (km) Ratio

Alphonsus 694 0.069-4 Arzachel 148 . .0253 Atlas 388 .0637 Balboa 135 .0397 Balboa "W" 103 .0981

Balboa "X" 11.8 .0141 Balboa "Y" 27.0 .1330 Blancanus C 95.4 .0815 Boscovich 60.2 .0409 Briggs 143 .1310 Byrgius D 106 .1790 Cardanus 68.5 .0357 Cassini 68.6 .0320 CrugerE 15.8 .0868 Dalton 189 .0591

Daniell 61.6 .1080 Einstein A 105 .0214 Gauss 845 .0356 Hansteen 188 .1130 He din "X" 65.2 .1780

Hevelius 476 .0482 Humboldt 1373 .0439 La Condamine 112 .1030 LaGrande W 179 .0886 Lavoisier 652 .1645

Lavoisier A 65.7 .1200 Lavoisier B 18.3 .0398 Lavoisier D 526 .1750 Lavoisier E 305 .1410 Lavoisier F 152 .1750

Lavoisier H 139 .1940 Lavoisier Q 12 , .1270 Mersenius 490 .0922 Messala 181 .0177 Petavius 313 .0118 170

Total Length of Cracks Cracked (km) Ratio

Pitatus 773 0.0974 Posidonius 181 .0242 Repsold 454 .0534 Repsold G 158 .1040 Riemann "X" 60.1 .0305 Stuve G 17.1 .0925 Tannerus 41.8 .0629 Thebit 242 .0975 Vasco da Gama 358 .0470 Vasco da Gama R 194 .0915

Vasco da Gama "V" 72.6 .2580 Vasco da Gama "W" 90.5 .0494 Vitello 131 .0936 Wurzelbauer 274 .0545 APPENDIX IV

CRACK PATTERNS

171 Anastomosing Intermediate- Concentric Concentric angle Subradial Radial Crater Cracks Cracks Cracks Cracks Cracks

Alphonsus - 3 - 3 - Arzachel - 2 - 1 Atlas — ■ 3 - 2 - Balboa 2 - - ■ - 1 Balboa "X" — — 2 1 -

Balboa "Y" 1 — — — 2 Blancanus C 3 - —— 1 Boscovich > -- 2 - Briggs 4 — - 1 2 Byrgius D 3 — 1 Cardanus 2 — 1 -- Cleomedes - i 1 1 - Cruger E -— 1 - - Damoiseau - 3 1 1 - Daniell 3 —— 2 — Doppelmayer 4 1 1 2 - Einstein A - 2 1 - 1 Galilaei B 4 - - 1 - Hansteen - 4 1 3 - Hesiodus 2 —— 1

Humboldt 4 — 1 - 4 La Condamine - 3 1 3 - Lagrange W 1 - 1 - 2 Lavoisier - 4 1 1 - Lavoisier A - 3 1 2 —

Lavoisier B — 2 1 — - Lavoisier D - 3 1 - 2 Lavoisier E - 4 3 2 - Lavoisier F - 4 2 1 - Lavoisier H - 3 4 —-

Lavoisier "Q" 3 - 2 1 - " Mare Smythii B" - 3 1 - " Mare Smythii C" - 3 - 2 - Mersenius - 3 1 3 - Messala -- 1 1 - 173

Anastomosing Intermediate Concentric Concentric angle Subradial Radial Crater Cracks Cracks Cracks Cracks Cracks

Pitatus 4 2 Posidonius 1 1 - 1 1 Repsold - 3 1 - 3 Repsold G 2 — 1 2 - - Thebit — 1 1 2 —

U. Beigh "X" — ' . 3 - -- Vasco da Gama " V" 2 -- 1 — Vitello - 3 - 1 - unnamed — 2 1 1 - Totals: 39 73 33 45 23

Average values 0.89 1.66 0.75 1.02 0 *52

Combined Concentric Average: 2.55 (Good-Fair) Intermediate Average: 0.75 (Very Weak) Combined Radial Average: 1.75 (Fair-Weak) REFERENCES

Adler, Joel M ., and Salisbury, John W. , 1969 , Circularity of lunar craters: Icarus, v. 10, p. 37-52.

Arthur, D. W. G ., Agnieray, Alice P., Horvath, Ruth A., Wood, C. A., and Chapman, C..R., 1963, The system of lunar craters, quadrant I: Univ. Arizona Lunar and Planetary Lab. Commun., v. 2, no. 30, p. 71 Arthur, D. W. G., Agnieray, Alice P., Horvath, Ruth A., Wood, C. A., and Chapman, C. R., 1964, The system of lunar craters, quadrant II: Univ. Arizona Lunar and Planetary Lab. Commun., v. 2, no. 40, p. 1-59. Arthur, D. W. G ., Agnieray, Alice P.,Pellicori, Ruth H„, Wood, C. A., and Weller, T ., 1965, The system of lunar craters, quadrant III: Univ. Arizona Lunar and Planetary Lab. Commun., v. 3, no. 50, p. 61-146. Arthur, D. W. G., Pellicori, Ruth H., and Wood, C. A., 1966, The system of lunar craters, quadrant IV: Univ. Arizona Lunary and Planetary Lab. Commun. , v. 5, pt. 1, p. 1

Baldwin, Ralph B., 1963, The measure of the moon: Chicago, Univer­ sity of Chicago Press, 488 p. Cameron, W. S., 1964, An interpretation of Schrftter's Valley and other sinuous rilles: Jour. Geophys. Research, v. 69, p. 2423-2430. Carr, Michael H ., 1966, The structure and texture of the floor of Alphonsus: Ranger VIII and IX. Part II of Experimenters' Analyses and Interpretations: NASA-JPL (Pasadena), Technical Report No. 32-800, p. 270-284.

Carr, W. J. , 1964, Structure of the Timber Mountain dome and caldera, Nye County, Nevada: U.S. Geol. Survey Prof. Paper 501-B, p . 16-19. Christiansen, Robert L ., Lipman, P. W. , Orkild, Paul P ., and Byers, F. M ., J r ., 1965, Structure of the Timber Mountain caldera, southern Nevada, and its relation to Basin-Range .structure: U.S. Geol. Survey Prof. Paper 525-B, p. 43-48.

Cruikshank, Dale P ., 1965, The origin of certain classes of lunar maria ridges: unpubl. M.S. thesis. University of Arizona, 68 p.

174 175 Fielder, Gilbert, 1965, Lunar geology: London, Lutterworth Press, 184 p.

Firsoff, V. A„, 1961, Surface of the moon: London, Hutchinson, 128 p. Gilvarry, J. J ., 1960, Origin and nature of lunar surface features; Nature, v. 188, no. 4754, p. 886-891.

Gold, T„, 1966, The moon's surface, jui The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N„, Menzel, Donald H., and O'Keefe, John A., eds.: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 107-124. Hagfors, T ., 1966, Review of radar observations of the moon, _in The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N ., Menzel, Donald H. , and O'Keefe, John A ., eds .: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 229-240.

Hartmann, William K., 1964, Radial structures surrounding lunar basins, II: Orientale and other systems: Univ. Arizona Lunar and Planetary Lab. Commun., v. 2, no. 36, p. 175-188.

Hartmann, William K., and Yale, Francis G., 1969, Mare Orientale and its intriguing basins: Sky and Telescope, v. 37, p. 4-7.

Hathaway, Allen W. , 1971, Lava tubes and collapse depressions: un- publ. Ph.D. disseration, University of Arizona

Kuiper, Gerard P ., 1966, The surface structure of the moon, _in The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N. , Menzel, Donald H. , and O'Keefe, John A., eds.: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 99-106.

Lowman, Paul D ., Jr., 1969, Lunar panorama: Zurich, Switzerland, Weltflugbild Reinhold A. Muller, 105 p.

Mack in, J. H„, 1969, Origin of lunar maria: Geol. Soc. Am. Bull., v. 80, no. 5, p. 735-748. Marjaniemi, Darwin K., 1969, Geologic history of an ash-flow sequence and its source area in the Basin and Range Province of south­ eastern Arizona: unpubl. Ph.D. dissertation, University of Arizona, 176 p. 176

McCauley, John F „, 1966, . Intermediate-scale. geologic map of a part of the floor of Alphonsus: Ranger VTII and DC, Pt. II, jn Experi­ menters' Analyses and Interpretations: NASA-JPL (Pasadena), Technical Report No. 32-800, p. 313-319.

McGill, George E ., 19 71, Attitude of fractures bounding straight and arcuate lunar rilles: Icarus, v. 14, p. 53-58. Moore, Patrick, and Cattermole, Peter J., 1967, The craters of the moon: London, Lutterworth Press, 160 p. Murase, Tsutomu, and McBirney, Alexander R., 1970, Viscosity of lunar lavas: Science, v. 167, no. 1491-1493.

Mutch, Thomas A. , 1970, Geology of the moon: Princeton, New Jersey, Princeton University Press, 324 p.

O'Keefe, John A. , 1966, Lunar ash flows, _in The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N ., Menzel, Donald H., and O'Keefe, John A., eds.: sponsored by IAU Commission 17 (the Moon) and NASA, God­ dard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 259-266. Pohn, H. A ., and Offield, T. W ., 1969, Lunar crater morphology and relative age determination of lunar geologic units. Pt. 1: Classification: U.S. Geol. Survey, Interagency Report: Astro- geology 13, 35 p. Quaide, William, 1965, Rilles, ridges, and domes—clues to maria history: Icarus, v. 4, p. 374-389.

Ronca, Luciano B ., 1965, A geological model of Mare Humorum: Icarus, v. 4, p. 390-395.

Ronca, Luciano B., 1966, Structure of the crater Alphonsus: Nature, v. 209, p. 182. Schumm, S. A ., 1970, Experimental studies on the formation of lunar surface features by fluidization: Geol. Soc. Am. Bull., v. 81, p. 2539-2552. Shorthill, R. W ., and Saari, J. M., 1966, Recent discovery of hot spots on the lunar surface, jn The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N. , Menzel, Donald H., and O'Keefe, John A. , eds.: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 215-228. 177 Smith, Robert L„, 1966, Terrestrial calderas, associated pyroclastic deposits, and possible lunar applications, jn The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N., Menzel, Donald H., and O'Keefe, JohnA.,eds.: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 241-258.

Smith, Robert L ., and Bailey, Roy A., 1968, Resurgent cauldrons, jn Studies in volcanology, Coats, R. R., ed.: Geol. Soc. Am. Mem. 116, p. 613-662.

Smith, R. L., Bailey, R. A., and Ross, C. S., 1961, Structural evolu­ tion of the Valles caldera, New Mexico, and its bearing on the emplacement of ring dikes: U.S. Geol. Survey Prof. Paper 424-D, p. 145-149. 8purr, J. E ., 1944, Geology applied to selenology. Vol. 1, The Imbrian Plain region of the moon: Lancaster, Pa., Science Press, 112 p.

Strom, Robert G ., 1964, Analysis of lunar lineaments. I—Tectonic map of the moon: Univ. Arizona Lunar and Planetary Lab. Commun., v. 2, no. 39, p. 205-221. Titley, S . R., 1967, Geologic map of the Mare Humorum region of the moon: U.S. Geol. Survey Misc. Geol. Inv. Map 1-495.

Trask, N. J ., and Titley, S. R., 1966, Geologic map of the Pitatus region of the moon: U.S. Geol. Survey Misc. Geol. Inv. Map 1-485. Urey, H. C. , 1966, Observations on the Ranger photographs, _inL The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N ., Menzel, Donald H., and O'Keefe, John A., e d s.: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 3-22.

Whitaker, E. A. , 1966, The surface of the moon, jn The nature of the lunar surface: Proceedings of the 1965 IAU-NASA Symposium, Hess, W. N ., Menzel, Donald H ., and O'Keefe, John A., eds.: sponsored by IAU Commission 17 (the Moon) and NASA, Goddard Space Flight Center, April 15-16, 1965; Baltimore, The Johns Hopkins Press, p. 79-98.

Whitaker, E. A., Kuiper, G. P., Hartmann, W. K., and Spradley, L. H„, eds., 1964, Rectified lunar atlas, Supp . no. 2 to the Photo­ graphic lunar atlas: Tucson, The University of Arizona Press, 30 p. 178

Wilhelms, D„ E„, Masursky, Harold, Binder, A. B., and Ryan, J. D., 1965, Preliminary geologic mapping of the easternmost part of the lunar equatorial belt, _in Astrogeol. Studies Ann. Progress Rept. , July 1, 1964 to July 1, 1965, pt. A: U.S. Geol. Survey open-file report, p. 45-53.

J

77.5W

LEGEND ■ Filling Materials

M PROBABLY OLD MARE MATERIAL; IT HAS A HIGH ALBEDO. IS WELL CRATERED.AND IS FLAT AND OCCUPIES T0P0- GRAPHIC LOWS. IT IS I 0UND IN III'! II WEATHERED DEPRESSIONS ON THE UPLANDS AND ALONG THE EDGES OF MORE RECENT MARIA.

m A FILLING MATERIAL OF INTERMEDIATE AGE ; IT HAS mmmm AN INTERMEDIATE ALBEDO AND IS FLAT. ALTHOUGH SLIGHTLY HUMMOCKY. IT IS FOUND AS THE FILLINGS OF CRATERS WITH CRACKED FLOORS AND ALSO A FEW IRREGULAR DEPRESSIONS ON 42.5N THE UPLANDS.

VZW/JL TYPICAL MARE MATERIAL m z m WITH LOW ALBEDO.LITTLE U CRATERING. AND SMOOTH TOPO­ GRAPHIC FORM. THIS MATERIAL IS YOUNGER THAN M, OR Mg AND ERARD, BELONGS TO THE PROCELLARUM GROUP.

VM4 MARE MATERIAL WITH 1 ' I VERY LOW ALBEDO. VERY 1 L IT T L E CRATERING. AND VERY SMOOTH TOPOGRAPHY. PROBABLY 6 1 A VERY YOUNG MATERIAL.

Miscellaneous

UNDIFFERENTIATED UPLAND MATERIAL WITH HIGH ALBEDO AND SMOOTH TO HUMMOCKY TOPOGRAPHY.

CRATER WALLS WITH SHARP RIMS : THE YOUNG­ EST CRATERS. 42.5N

CRATER WALLS WITH UNEVEN AND DULL RIMS THAT APPEAR TO HAVE BEEN WEATHERED OR AGED. THESE ARE THE OLDEST CRATERS. 0 CENTRAL PEAKS . EJECTA BLANKETS, AND DISTURBED UPLAND MATERIALS.

- . - STRUCTURAL LIN EA R FEATURES' PROBABLY FAULTS OR FISSU R ES.

37.5N CRACKS BELONGING TO CRATERS WITH CRACKED FLOORS.

GRABEN TYPE FAULT # WITH FLAT BOTTOM; THIS FEATURE OFTEN CUTS UNDE - ELECTED ACROSS RIDGES AND LAVOISIER VALLEYS.

m CRATER CHAINS

MARE WRINKLE RIDGES

©

77.5 W 62.5W 80 tV Figure 73. The Lavoisier Group of Lunar Craters

AREA MAPPED IS BASED ON THE LUNAR ORBITER ISC HIGH RESOLUTION PHOTOGRAPH, FRAME 189, 2nd of 3 parts, SHOT ON MAY 25,1967.

SCALE IS APPROXIMATELY I INCH EQUALS 10.4 MILES OR I CENTIMETER EQUALS 6.6 KILOMETERS.

R.N.WELLER MAPPED BY R. WELLER M.S.THESIS GEOSCIENCES 1972 NORTH

Figure 67.

Map of the Frontside of the Moon Showing the Distribution of Craters with Cracked Floors %

NOTE' EQUAL AREA PROJECTION HAS BEEN USED TO MINIMIZE DISTORTION OF THE LIMB REGIONS.

SCALE IS APPROXIMATELY 188.5 STATUTE MILES OR 304 KILOMETERS FOR EACH 10" SERE OF LATITUDE.

I SQUARE INCH IS APPROXIMATELY EQUAL TO 2 6 ,5 0 0 SQUARE MILES OR I SQUARE CENTIMETER IS APPROXIMATELY EQUAL TO 10,640 SQUARE KILOMETERS.

LEGEND

WELL-FLOODED MARIA; BOTH CIRCULAR AND IRREGULAR

REGIONS OF SHALLOW FLOODING OF MARE MATERIAL iftfir# 0 CRATERS WITH V.e CRACKED FLOORS

MAPPED BY R WELLE* 0 9 1 * 0 THE NATIONAL OEOOMPMIC SOCIETy'9 MAP OF THE EARTH 9 MOON AS A REFERENCE

SOUTH

P M W FI I F D TUFCIC