Role of Seedlings in Recovery of Midwestern Ranges from Drought

Home , Diaperia

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

Agronomy & Horticulture -- Faculty Publications Agronomy and Horticulture Department

7-1942

Role of Seedlings in Recovery of Midwestern Ranges from Drought

J. E. Weaver University of Nebraska-Lincoln

Irene Mueller

Follow this and additional works at: https://digitalcommons.unl.edu/agronomyfacpub

Part of the Plant Sciences Commons

Weaver, J. E. and Mueller, Irene, "Role of Seedlings in Recovery of Midwestern Ranges from Drought" (1942). Agronomy & Horticulture -- Faculty Publications. 451. https://digitalcommons.unl.edu/agronomyfacpub/451

This Article is brought to you for free and open access by the Agronomy and Horticulture Department at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Agronomy & Horticulture -- Faculty Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. Ecology, Vol. 23, No. 3 (July 1942), pp. 275-294 Copyright 1942 Ecological Society of America. Used by permission.

ROLE OF SEEDLINGS IN RECOVERY OF MIDWESTERN RANGES FROM DROUGHT * 1

J. E. WEAVER AND IRENE M. MUELLER University of Nebraska

An extensive survey of the degree of would not be sufficient to improve greatly deterioration of midwestern ranges was the range as regards the growth of new made in 1939 (Weaver and Albertson, plants. Exceptions are buffalo grass '40). Decreases in the basal cover of the (Buchloe dactyloides), western wheat grasses from extreme drought, overstock- grass (Agropyron smnithii), and a few ing, and damage by dust were appalling. other species which have excellent means Basal cover of the native grass had de- of propagation by stolons or rhizomes. creased to 1 per cent in one-fourth of the Consequently, it was planned, early in ranges examined and it was 21 per cent the spring of 1941, to make a survey of or more in only 16 per cent. Little, if the viable seed population in the surface any, improvement occurred during the soil of these depleted grasslands. This following year; instead certain large areas survey included also drought-damaged suffered distinct losses as a result of true prairie of eastern Nebraska and severe late summer drought. The peren- Kansas. Here, where the mean annual nial grasses over large areas in south- precipitation is 20 inches or more, in- western Kansas, for example, had been crease in abundance and vigor of vegeta- reduced to a basal cover of about 3 per tion since the early years of drought ('34- cent in 1937. The cover improved slightly '37) has often occurred, but recovery is during the two following years, but was usually far from complete (figs. 4 and 5). reduced to less than 0.5 per cent during Later a quantitative study of the seedling the very dry fall of 1939 (Albertson, population was also made. Fortunately, '41). On some of the better ranges near 1941 was a year of abundant rainfall and Syracuse, where the basal cover was about consequently very favorable to seed 15 per cent, scarcely a spear of grass re- germination. mained (figs. 1, 2 and 3). The drought has now continued (to COLLECTING SAMPLES IN THE FIELD 1941) with only slight interruptions for Seeds of range grasses, with rare ex- eight years. Occasional showers have ceptions, are found on or near the sur- moistened the soil sufficiently at irregular face of the soil; therefore samples of soil intervals to promote germination of seeds. were taken to a depth of 0.5 inch only. including any that may have lain dormant Each sample included the top soil of an for a long time, but rarely has there been area of one square foot. It was secured establishment and continued growth of by means of a sharp, flat-bottomed shovel Production of viable seed by seedlings. 10 inches in width, which was pushed the native grass has usually been pre- through the surface soil at 0.5 inch depth vented weather following too by dry to a distance slightly greater than 14 rapidly a period of flowering. Conse- inches. Samples were collected from 49 quently, it has seemed that the seed sup- stations in 28 counties. Except for 13 ply in the soil might have become greatly counties in western Kansas, where 3 sam- depleted and that a single growing season ples were taken at each station, five * Contribution from the Department of Bot- square-foot samples were regularly ob- any, University of Nebraska, No. 133. tained from each pasture or range. Sam- 1 This study was made with the aid of a grant from the Penrose Fund of the American Philo- ples were placed in separate paper sacks, sophical Society. taken to the laboratory, and air dried. 275 276 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3

FIG. 1. Range near Syracuse, Kans., showing almost complete destruction of vegetation. Only a few bunches of blue grama grass remain. Photo, Au- gust, 1939. FIG. 2. Pasture near Dighton, Kans., showing cracking of soil in layer of dust and rosettes of winter annuals. Photo by Albertson, April 11, 1941. FIG. 3. Detail of range near Cheyenne Wells, Colo., showing small tufts of blue grama and much bare soil. Typical of vast areas in the Great Plains. Photo, June 17, 1941. July, 1942 ROLE OF SEEDLINGS 277

FIG. 4. Prairie near Lincoln, Nebr., with new growth from prairie dropseed and needle grass which had previously been clipped. Note the large amount of soil without cover. This is representativeof one of the better drought-damaged prairies. Photo, August, 1940. FIG.5. A good range near Idalia, Colo. Chief species are blue grama, sand dropseed, and buffalo grass. Note lack of debris on the hard, bare soil. The wide range of sampling included 7 (Lincoln B, Nebraska, Jennings A, Kan- counties in Nebraska, 4 in eastern Colo- sas, and Stratton, Colorado) on April 16. rado, and 17 in Kansas (fig. 6). Alto- The remaining samples were kept air dry gether, 196 samples were obtained. Col- until May 27 or 29. Benches in the lections were made either late in March greenhouse were cleared and a shallow or early in April at a time when no grass layer of fine gravel was laid in the bot- seedlings were found, although some of toms to provide drainage. A layer of the winter annuals, as Lepidiurn densi- sterile sand about 2 inches thick was florum and Hordeum pusillum, were an placed over the gravel, after which the inch or more in height. planting area was divided into square- foot plots by lath partitions. Addition of STUDIES ON GERMINATION more sand brought the layer within 0.75 Studies on germination were begun inch of the top of these partitions. Each with samples of soil from 3 stations sample of soil was placed in a separate 278 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3

PHELPS RIVER MORE ~~~LAN-

ADAMS C-LAY FIL ALINE4 CASTER NEBRASKA

IHARLAN FRANK- WENSTE NUCK- THAYER LIN N OLLS F _ __ _ L~~~ - ? ---_ __hs_ DEI AUT | 1CATUN NORTON I PHILLIPS-l

YUMA1 25

LIN- KIT CARSON ! SHERMAN THOMAS SHERIDAN GRAHAM ROOKS COLN RN KANSAS

WALLACE LOGAN GOVE' SO TREGO ELLIS CHEYENNE 3425332

GREELEY WICHITA SCOTT I LANE NESS RUSH 36 42 47 4946 I ~~~~~~~43 45 UN COLORADO

HAMiWTON KEARNEY FINNEY

~41I

FIG. 6. Countiesand station numberswhere studies were made. Names of stations are given in table I. Holdrege (10), Alma (11), and Phillipsburg (25) are among the most easterly stations in mixed prairie. plot, spread to a uniform thickness over house while the study was in progress. the sand, and pressed into firm contact Panels of glass from the end of the with it. A depth of planting exceeding house were removed and the opening 0.5 inch to 1 inch has been found detri- covered with woven wire, allowing free mental to most grasses (Clements and circulation of air. Light was necessarily Weaver, '24; Savage, '34; E. C. Con- reduced somewhat when alternating bands ard 2). Consequently, the soil layer was of whitewash were applied to the roof, a limited to 0.5 inch to insure germination procedure which seemed advisable since and emergence of all species. The par- it resulted in lower day temperatures. titions extended about 0.25 inch above the Records from a hygrothermograph, op- soil after settling, and thus prevented it erated at about the same level as the plots from washing from one plot to another of soil, showed that prior to July 21, day when watered. The plots were sprinkled temperatures rose above 1000 F. only 6 with a fine spray of water as often as times. After that, however, the daily there was evidence of drying. The ex- maximum temperatures were regularly periment was continued until new seed- above 1000 F. Minimum temperature lings appeared only rarely. About July 1 never fell below 550 F. Evaporation of the soil was allowed to dry, and each sam- water from tubs and frequent sprinkling ple was lifted into a large pan where it of the floor kept a normal amount of was pulverized and thoroughly mixed. water vapor in the air. Relative humidity The samples were then returned to their was a little below 30 per cent only twice. proper plots and treated as before until More often it ranged between 40 and 80 the end of July. per cent. An environment favorable to germina- Seedlings were counted, recorded, and tion of seeds was maintained in the green- removed from the plots as quickly as 2 Master's thesis, University of Nebraska, they could be identified. Consequently, 1938. Unpublished. none died from competition. Some of the July, 1942 ROLE OF SEEDLINGS 279 common ruderals were readily recognized made before 3 or more leaves had devel- as early as the cotyledon stage but usu- oped (fig. 7). This is in accordance with ally 2 to 3 leaves had developed before Copple and Aldous ('32) who state that the plants were removed. Several of the the earliest period at which identification less common species were allowed to by vegetative characters appears practical flower before identification could be cer- is after the production of the second blade. tain. Most of these plants, especially When distinction between two species was from western stations, attained maturity difficult, plants were carefully lifted to so quickly that identification of all but 3 secure the caryopses (sometimes burs) species of minor importance was possible which were still attached to the seminal when the study was concluded at the end roots. of July. Results Ordinary manuals are of little value for the identification of grass seedlings. Seedlings of Russian thistle (Salsola Several bulletins and leaflets on vegetative pestifer) and pigweeds (Amaranthus characters of pasture grasses were used spp.) were the first to appear in the plots, (Copple and Aldous, '32; Keim et al, and germination of these continued until '32; Harrington, '38). These were help- mid July. The prostrate spurges (Cham- ful sources of information, but check aesyce spp.), purslane (Portulaca oler- seedlings of most of the common prairie acea), lamb's quarters (Chenopodium and pasture grasses, grown in boxes in spp.), stickseed (Lappula occidentalis), the greenhouse, were the most reliable and peppergrass (Lepidium spp.) also means of identification. Although dis- germinated promptly. Seedlings of most tinguishing characters appeared relatively grasses emerged at the same time or early in grasses, verifications were not shortly after the ruderal forbs. Germina-

FIG. 7. Seedling grasses three weeks after planting the seed. Left to right in upper row are little barley (Hordeum pusillum), little bluestem (Andropogon scoparius), western wheat grass (Agropyron smithii), side-oats grama (Boute- lobaacurtipendula), and tall dropseed (Sporobolus asper). Lower row: witch- grass (Panicum capillare), big bluestem (Andropogon furcatus), tumblegrass (Schedonnarduspaniculatus), and buffalo grass (Buchloe dactyloides). Plants grown out-of-doors under very favorable environment in July. r 1 - 1 cl a) ce I Cu,~e> croc m cl r-c, oo eo.r o~~~~~~~~~~~~~~~~~~ ; o I N u~~~~~~~~~~~~~~~~L)C, em oo C eM

v - - I 0t ., "O\ oo O" XQ~~~~~~~~~~~~~~~~~~~~~tC sSNc0 0 u oXeorN X ~~~~~~~sqlojpe1lol + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~a)llqoe

t:3; t~~~~~~volnutffso-o-viuvid|

o = ~~~~~ddsutntpo~oudyD) _ _ _ U~~~~~~~v-~ppgo

_ txS ~~~~~vsoajqviqvuaq,'Aa

^ l_ ~~~~vuvfillvilugt~ajouoNr

t v rn | % > ~~~~~~~~~snsotuviuoiaXtiy _ X k ~~~~~~dds utntp!4.97 e

vaqvlR1o vgvolnlio |

,r tj0 0 I)0 t-t e

< < ) _ uX _ fA t _ - - N C c~~~~~~~~~~~~~~~~~~~~~nsIt 1 t- C14 mC> 4 \ t;; ~~~~(al-ellsold) 'dds -qq1s-qvutvl, |>DN q OrC) 0s UO U)L C0e C1 14 0 XU JK1S~~~~~~~'aff VOSya_W

t _ 71 - . LI U) enC0,4 e 0 ^ = e e N W O N m O e X + N~~~~vt--0 t.? dds snyluvmsut eno c

4- - u- en "t o ~cU, , 4 o I C1 \0 C, C1 Ln Ln LO u,) -d 't t n eN > en _O e It sassel. pel?L>_

ewQ ~~~~~~~sassel.5laqlo e+ :

AZ; ~~~~vjnpuq4f!jmtvnopl9no |g

. < ~~~s-njvjnlnauv s-npivuuop~ll |oos

;_, X D~~~~~~~~~~1)1St,19 Vl|OX| Q $ ~~~~~~~dds XgivD| ___ = olO ~~~~Sgpolxlg1Vp golqqng |XNt (Zf) 0 __ Y V vs-ouvnbs- voiunN

o ts 0 0 ~~~~~~~~~~uznlltsn4fuMngp-iom .:; - \s I!, S? ~~~~~~~~StltDA?Dno191nog

C' onC' \0CbS 00-e4 C' cq

n ,, 4 l)snfpuvjffiq snloqoioff gbo A0 E

O; U )~~~~~jbs 'saldmieS (m m m mm m me s E :^~~~~~~~~~~ . ~. . ~~~~~~...... (n .bZ

_~~~ ~ ~ ~on ~ ~~~~~~~~~~~~~~~~~~a''= U;=Qh0Esaccbl Cn V sa 0 C ...... cn ...... =t _csdLorO-MWr~bce Cd 0. 4v...... C)..C)... Ni o O to .d "t t- cs 't \0 0 - \0 C cs "I M M \O cs - M O 0

O~~~~~~~~~~~~~~~~~CIq c s Iscs 0 0o

oo r oo oo ,d cs to CS4 cq U) C, \ oo -t C, \ M r- 00 o -4\ X sq~iojItl? oL

sq~ioj iaqlo I IVX

vsolnut$SOs- vlcts8~ f |

-dds ullutpoffouglio| I

sviv2Xpil~n o.?n2llop | t IXo

vso~glvvxqvusqqS |>9

r- vouvtllvlnu s-t?ajouopv | cs _ ? _Nt-

o ~~~~s-ns-ouiviuoig.ti2 | o

X ~~~~~dds utnip!4;7 \ s 4\ o Co vggv~~igl_vgln o -4 \ o0

slvjugp2q vlvn4oD7 c bXc

- v r- >N M C, \ 00 Ch+4 \se C1 rr -dds;qx-qlvJ ~

LO t :o cM \ CS 00 CS4 C*Ido t- CS4 V) \ CS r- It cso 1qf}S-4 v20sIDS 1-4~mu~mo + o

,dds s-nyuvivutF : s er >~ ~~ ~ ~~~~~~~~~~~~rI. LnC4L Xr 1 ,- q 1 qr qU)C0Mt

l ~~~vjnpug4.tnq vnoldinog |IT

Xss~~~~.tuzjnant2 sncq4X,3| c

s-njvjnnuv4fs-npivuuop2VgS |

bO ~~~~~dds xgivo)|c

S<1ollv -oyn MJ NN N N - W _4 ICo tl- \10 s) M 00 r-_ sXvsoiivnbs votunN \, csqo

gOivjpfvq uwnatuvi |0 c oo

uwnllssn4 utngpioH LOU cs cs

M0C)C \ T CSU O N 4N+

EJ t X - N- N X N X N IX N IJ CN O )-4- \0 00 t- _4 O O -, t- C, C, in 00 \0 -4 cq M \ o StSU9U~tj!9 911SOIlDiV

N XE N N _ _ \ Cq r _ s-nxpuvj4Xa.s-nloqoioff c o s +e *Ij*bs aldwet s+ 4

V) , X * ...... ,.SM-) C! X . - .' a

oroo~e;+meroo~e+Worw w .v.v )U O c0 0 C)e+d+++d< C) 282 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3 tion of stinkgrass (Eragrostis cilianenisis) dactyloides), all recognized as valuable was delayed several days, however, and forage species. This was an average of that of little barley (Hordeum pusillum) 25.1 plants per square foot. Sand drop- for about 3 weeks. The only grass seeds seed alone made up 63 per cent of the which germinated abundantly after the grass plants and occurred in 114 samples soil had been worked over were sand f rom 88 per cent of the stations. Blue dropseed (Sporobolus cryptandriis), little grama grass (Bouteloua gracilis) ap- barley, and stinkgrass. peared in 85 samples and was missing More than 13,000 seedlings emerged from less than one-third of the stations. from the 196 samples of soil, an average Buffalo grass, recorded from only 25 per of nearly 67 plants per square foot (table cent of the stations and 11 per cent of I). The total number of grass plants the samples, lacked the uniform distribu- (6,730) somewhat exceeded the total tion of the two preceding grasses. Ex- number of forbs (6,388), but the num- cept for a small number of seedlings of ber of species of forbs (40) surpassed June grass (Koeleria cristata), side-oats the species of grasses by 14. Only 2 grama (Bouteloua curtipendula), big samples, both from a severely dusted bluestem (Andropogon furcatus), little pasture at Norton, failed to produce any bluestem (A. scoparius), needle grass seedlings. Germination from samples col- (Stipa spartea), and sedges, the re- lected at Carleton, Crete, Hays A, Lin- mainder consisted of weedy species. coln A, and Pleasant Dale were also rel- Samples from Stratton, Norton, Tribune, atively low. Seed failed to mature at Holcomb, and Ness City B failed to pro- these stations in 1940. Samples f rom (luce forage grasses in significant num- Laird yielded the greatest number of bers. plants, 2,062, an average of 687 per square Sixteen forbs-Lappitla occidentalis, foot. About 68 per cent of these were Portulaca oleracea, Monolepis nuttalliana, sand dropseed. Plantago spinulosa, Talinum parviflorum, Certain species of grasses and a sedge Neomamillaria sp., Opuntia humifusa, -Koeleria cristata, Bouteloua curtipen- Diaperia prolifera, Psoralea tenuiflora, dudla,Andropogon scoparius, A. furcatus, Oxalis stricta, Specularia leptocarpa, Al- Stipa spartea, Eragrostis pectinacea, lionia linearis, Proboscidea louisiana, Si- Echinochloa crus-galli, Agrostis hyemtalis, symbrium altissimum., Sophia pinnata, Panicum scribnerianum, Bromus tecto- and 1 unknown species except 1 speci- rum., and Cyperus filiculmtis-were more men each of Lappula and Portulaca, were or less limited to the true-prairie stations limited to mixed prairie. Only 9 species (Pleasant Dale, Lincoln, Crete, Clay Cen- -Aster multiflorus, Hosackia amvericana., ter, Fairfield, Carleton, Nelson, and Su- Specularia perfoliata, Petalostemon sp., perior). They exceeded the species com- Trifolium repens, Veronica xalapensis, mon to mixed prairie-Buchloe dactyl- Silene antirrhina, Mentha sp., and 1 un- oides, Schedonnardus paniculatus, Mun- known-were confined to true prairie. roa squarrosa, Bouteloua hirsute, and one Amaranthus spp., Salsola pestifer, Chain- unknown-by 6. Sporobolus cryptan- aesyce spp., Lepidium spp., Erigeron drus, Eragrostis cilianensis, Bouteloua ramosus, Verbena bracteosa, Mollugo gracilis, Horde inn pusillum, Panicum verticillata, and Chenopodium spp. were capillare, Festuca octoflora, Setaria spp., recorded from samples over the whole and Carex spp. were common to both area. The scarcity of seedlings of native regions. perennial f orbs was noteworthy. More Seventy-three per cent of the total than 96 per cent of the total f orbs con- number of grass plants consisted of 3 na- sisted of annual weeds. In general, more tive perennial grasses (Sporobolus crypt- weed seeds germinated in samples f rom andrus, Bouteloua gracilis, and Buchloe west-central Kansas than from any other July, 1942 ROLE OF SEEDLINGS 283

area, but the production of weeds was TABLE II. Minimum and maximum depth in inches to dry soil when samples for germination also high in samples from Fairfield, which studies were collected in spring and when seedling is a true prairie station. surveys were made in early summer. In location of stations by states, Nebraska, Kansas, and Colorado are represented by N., K., and C., FIELD STUDIES respectively. Station numbers correspond to those on the map (fig. 6). The ranges from which soil samples for seed germination had been secured Mar. 28- June were studied between June 10 and 26. Station Apr. 12 10-26 The soil was examined in regard to water 1. Pleasant Dale, N.. 28-40 30-40 2. Lincoln, N ...... 24-50 36-60 content and consequent suitability for 4. Crete, N.. 21-48 30-48 seed germination, the composition and de- 5. Clay Center, N ...... 25-29 25-29 velopment of the vegetation were studied, 6. Fairfield, N ...... 11-13 18-60 7. Carleton, N ...... 23-24 42-46 and the abundance of seedlings was ascer- 8. Nelson, N.. 13-22 30-35 tained. 9. Superior, N ...... 11-27 40-42 10. Holdrege, N.. 10-11 11-24 11. Alma, N. 16-18 18-24 Soil moisture 12. Norton, K ...... 16-21 12-16 13. Jennings A, K ...... 21-27 The depth of moist soil had been deter- 14. Jennings B, K ...... 22-26 35-42 mined at the time the samples were taken 15. Colby, K ...... 9-10 22-23 16. Brewster, K ...... 7-10 20-43 and a second survey of the depth of water 17. Goodland, K ...... 9-15 18-24 penetration was now made. On both 18. Kanorado, K ...... 12-16 18-27 visits, samplings were made by means of 19. Burlington, C...... 11-13 24-31 20. Stratton, C...... 5-11 21-22 a Briggs geotome at three different places 21. Seibert, C. 12-13 17-18 in each range. The deeper soil was so 22. Idalia, C.. 12-14 17-20 dry that the transition from moist to dry 23. Cheyenne Wells, C...... 6-12 25-26 24. Arriba, C...... 18-20 24-32 soil was very sharp and readily determined (table II). Average depth of moist soil, in inches, sas. The spring was cool, late, and, like at stations in west-central Kansas (num- early summer, unusually wet. The total bered as on map, fig. 6) was as follows precipitation was not only well above the on April 10 to 15: (25) Phillipsburg, 17; normal at every station, but often several (26) Stockton, 24; (27) Hays A, 23; inches above, and not infrequently two (28) Hays B, 22; (29) Ogallah, 22; to four times normal. Moreover, the (30) Quinter, 18; (31) Buffalo Park, showers were so well distributed that soil 18; (32) Winona A, 9; (33) Winona B, moisture was unusually favorable for 9; (34) Wallace A, 9; (35) Wallace B, germination of seeds and establishment of 9; (36) Tribune, 7; (37) Syracuse A, seedlings. Numerous days with a trace 18; (38) Syracuse B, 18; (39) Syracuse of rain or light precipitation indicated C, 21; (40) Holcomb, 30; (41) Kalvesta, much cloudy weather. No prolonged 18; (42) Dighton A, 20; (43) Dighton periods of drought occurred. Although B, 17; (44) Dighton C, 18; (45) Laird, the second survey of depth of soil mois- 22; (46) Ness City A, 17; (47) Ness ture was not made at all stations in west- City B, 20; (48) La Crosse A, 20; and central Kansas, repeated observation at (49) La Crosse B, 20. the numerous stations showed that mois- These data have been supplemented by ture was available for growth during the a study of the precipitation by days from entire spring and-early summer. April 1 to June 15. The study was made These data showed conclusively that on from the records of the Weather Bureau practically every range a continuous sup- at stations near the areas sampled. ply of moisture had been available to pro- Twelve such stations were examined in mote germination of seeds and establish- Nebraska, 5 in Colorado, and 20 in Kan- ment of seedlings, a fact which was borne 284 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3 out by the excellent development of weeds straw-colored, older ones a drab gray. and other vegetation. Great alternes on the barer slopes were composed of this weed, and in places they Composition and development of covered the entire pasture. Frequently vegetation 160 plants grew on a single square foot. Peppergrass was fairly abundant on the Depleted ranges are usually heavily in- more easterly ranges, grew thickly in fested with a diverse population of weeds. more central ones, but westward it was In spring especially, they give the appear- replaced on the bared areas by species of ance of excellent pasturage. Since they Lappula or Cryptantha. greatly affect the establishment of seed- While young plants of stickseed (Lap- ling grasses, mention will be made of the pula occidentalis mostly) were green, they more important species. covered the bared soil. On dried plants The annual (often winter annual) little the leaves had shriveled or fallen and the barley was dry and yellowish by June 15. much-branched stems and spiny fruits Dense, continuous stands 3 to 15 inches only were prominent. Their drab color tall gave many in ranges Nuckolls, Har- demarked many areas of a few square lan, Phillips, and adjoining counties the feet to many square rods, if indeed it did appearance of a stubble field or of headed not give tone to the whole landscape (fig. grain. In western Kansas it was found 11 ). Beneath this thick weedy growth, on the more favored sites only, and dis- usually about 8 inches high, was the de- appeared almost entirely in Colorado. pleted stand of grasses, the density of the Little barley is very detrimental to the grass cover usually being inversely pro- growth of grass seedlings and established portional to the density of the stand of short grasses as well. Like other rapidly the weeds. Stickseed was especially abun- growing weeds, it absorbs large amounts dant in western and of water from the surface soil and pro- Kansas Colorado. duces a thick layer of vegetation which The harsh borage (Cryptantha cras- reaches a height of 10 to 16 inches (fig. sisepala) is often associated with Lappula 8). The shade is dense, and the large or alone covers much bared or partly amounts of dried straw continue to ob- bared soil. These early maturing plants struct the light for seedlings and all low take on a blackish color when they dry. growing vegetation. When it lodges, All of these quickly maturing, weedy chance for survival of seedlings beneath plants-little barley, peppergrass, stick- it is still further reduced. At Phillips- seed, and cryptantha-frequently grow burg, Kansas, in 1940 and 1941 for ex- even in the best remaining patches of blue ample, this weed alone produced 1 to 2 grarna and buffalo grass, as well as more tons of forage per acre each year. densely between them. They form either Peppergrass (Lepidium densiflorum) a discontinuous, patchy layer above the formed continuous dense stands over grasses or overshadow them entirely many acres. The plants were about 8 throughout the entire range. They are inches tall and grew so thickly that they often accompanied by an abundance of obscured the grasses beneath. Density of six-weeks fescue (Festuca octoflora). stand always increased with sparseness of They are of little value for grazing and the perennial grass cover, which was of- upon drying become highly undesirable. ten only 1 to 3 per cent (figs. 9 and 10). Yet they usually formed the major com- The fallen seeds were sometimes so abun- ponents of the plant cover. Studies on dant that they completely covered the soil. the yield of ranges in western Kansas, Often the leaves of the grasses were splat- have shown that while a ton per acre of tered with the seeds which are mucilagi- peppergrass, stickseed, Russian thistle, and nous when wet. Newly dried plants were other weeds was produced, the yield of na- July, 1942 ROLE OF SEEDLINGS 285

FIG. 8. Little barley 10 inches high in overgrazed pasture at Lincoln, Nebr., June 7, 1939. FIG. 9. Peppergrass 8 inches high near Alma, Nebr., June 16, 1941. FIG. 10. Typical understory of blue grama grass growing under a dense cover of peppergrasswhich has been removed by pulling up the plants. Chey- enne Wells, Colo. FIG. 11. Dense cover of stickseed (Lappula occidetstalis) 8 inches high near A'rriba,Colo. A two per cent cover of short grass occurred beneath. Photo, June 19, 1941. 286 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3 tive grasses was only a few pounds.3 Responding to the excellent condition \Veeds were abundant generally, but on of soil moisture, crop plants had made some Colorado ranges they were few their best development in 15 years, and (fig. 12). roadside weeds were correspondingly The preceding weeds are all well thrifty. Stolons of buffalo grass were adapted to a dry grassland climate where spreading widely where this species re- moist springs are frequently followed by mained; blue grama was tillering abun- early summer drought. The dominant dantly, and its flower-stalks and spikes short grasses and mid grasses are simi- were so plentiful as to give a dark or al- larly adapted. Flower stalks of Bouteloua most black tone to large intermittent gracilis were often 13-15 inches tall by patches in the landscape. On tufts of the middle of June and Buchloe dactyl- blue grama only 5 to 6 inches in diam- oides had flowered and was developing eter, 125 to 150 flower stalks had grown, seed. Various species of wire grasses so favorable was the environment. Light (especially Aristida purpurea) were blos- yellow patches of flowering staminate buf- soming where they occurred in the sub- falo grass showed plainly at a distance, sere, and sand dropseed was beginning to often alternating with blue grama. blossom. Except for the great abundance of The spring had been too wet and cold weeds and the taller foliage of the peren- for the usual development of Russian nial grasses, vegetation had changed but thistles. They were greatly injured by little from its greatly deteriorated condi- late frosts and their growth delayed by tion observed in the fall of 1939 (Weaver cool rainy weather. In most pastures and Albertson, '40). The most important they were few, and in all they were sub- grasses on the midwestern ranges were ordinate to the other pasture weeds. blue grama, buffalo grass, and, in places, 3Albertson and Weaver, 1941. Unpublished sand dropseed. In comparison, other manuscript. grasses occurred only sparingly. In 1939,

0% ~ ~ ~ ~ ~ ~ 7

FIG. 12. South slope in a range with few weeds near Burlington, Colo. The bared soil has been naturally reseeded by sand dropseed. Most of the seedlings visible are one or two years old, although young seedlings occurred at the rate of 172 per square foot. Two parent plants are shown in foreground. This was the best reproductionfound in any of the ranges examined. July, 1942 ROLE OF SEEDLINGS 287

-_ R5

FIG. 13. An excellent ungrazed range of blue grama and buffalo grass in eastern Colorado. Note that the soil has been eroded about the individual bunches. Ruderals are not abundant. Photo, June 19, 1941. about 10 native perennial f orbs could be very little had been formed during the found on most ranges; some were fairly past eight years of drought. Nearly common. In this survey, perennial forbs every bunch or tuft of grass had 0.5 to were rare, except sometimes in low 1.5 or more inches of top soil eroded from ground and in subseres. it (fig. 13). There was neither mulch In southwestern Kansas, the fall nor shade for seedlings where weeds were drought of 1939 killed practically all few, except those that grew near estab- grasses over thousands of acres. Al- lished plants. There was abundant room though the ranges recovered slightly in for seedlings in all pastures and in all 1940, they were in no better condition at parts of most pastures. Moreover, con- the end of that year than when described ditions for growth had been so favorable in 1939 (Weaver and Albertson, '40). that no dead seedlings were found. Although the midwestern ranges appeared greatly improved in 1941, because of the Survey of seedling population abundance of foliage, many of them had Determination of the number of seed- only 1 to 8 per cent cover of perennial lings (June 10 to 26) could not have been grasses. Any improvement since the sur- well delayed until late in July, since it vey in 1939 amounted to the recovery would have been impossible to distinguish made in the spring of 1941. Stock was with certainty the previous year's seed- still excluded from many ranges; nu- lings from those of 1941. The survey merous other ranges were grazed only was made by means of determining the lightly. The soil surface was invariably number and kind of seedlings in each of bare and hard, since sandy lands were not 25 square-foot areas taken at random on included. Nearly all debris had been each range. Selection was made by swept away by wind and water; in fact, throwing a hoop, care being taken to in- 288 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3

elude samples from each of the different and individuals from true prairie to types of vegetation, from areas differently mixed prairie. Seedlings of the follow- grazed, disturbed by dust burial, and with ing were found only in the true prairie different slopes, etc. area: Koeleria cristata, Andropogon sco- The results of this survey are given for parius, Bouteloua curtipendula, Panicurn all of the stations, except those in 13 scribnerianum, Cyperus filiculmis, Pani- counties in western Kansas, in table III. cum wilcoxianum, Andropogon furcatus, Grasses which appeared in very early Stipa sparteac, Sporobolus heterolepis, spring and soon matured-Hordeurn pu- Bouteloua hirsute, Agropyron smithi, and sillum, Festuca octoflora, and Bromus Poa pratensis. tectorum-were not included in this cen- No species occurred in mixed prairie sus. Panicum capillare, Eragrostis cili- only. Seedlings of four species-Spo- anensis, and a few other grasses of a robolus cryptandrus, S. asper, Bouteloua weedy nature but longer lived were in- gracilis, and Buchloe dactyloides-were cluded with the native perennials. Seed- found more commonly in this western lings of native forbs were rarely found; prairie. Those of Schedonnardus panicu- they are not listed. Actually they played latus ranged throughout. only a small part in the regeneration of A total of 24 species of seedling grasses the range. Other grasses include a small were found in the true prairie stations but number of species found infrequently in only 7 in those of mixed prairie. Only true prairie only. two of the stations in true prairie-Fair- Examination of table III shows a de- field and Superior-were grazed, the cided decrease both in number of species others were drought-depleted, ungrazed

TABLE III. Seedlings of grasses and sedges from 25 square-foot samples from each of 22 ranges in southern Nebraska (N), northwestern Kansas (K), and eastern Colorado (C), in June, 1941. Num- ber of samples in which each species occurred is shown by superior numeral. The eight stations above the heavy line are in true prairie; others are in mixed prairie. Stations are shown by number in figure 6.

. a Station 6 N .. 2 8 5 2 a o

5. Clay Center,N...... *...- . 25 623,6218 4 l4ll 6

10. Holdrege, N ...... 25162123511C221301111 11. Alma, N ...... 27l~4N255179126 t |||||||5 17. CoodlayCente, KN...... 25 628312 623 a1 41 84 42 1 44 1 1 1 1 0 18. Fairfild, N ...... 25 2313193410142961a266 a 2 221. 6 38 14. Pleasants DaeK. 25 3 206 112 282711 8 895 43 f |1 171 1 12 53 32 194 15.21.Seiert LiColby, C.N ...... 25 25 35 2211 850 8919 311 19 '41 31 21 | 71 43 21 3 6490~ 512 4, '5964. 104 16.22. Crewter, dali, N ...... C. . 252 1 1 7 3411 4113 42 1 1 1944811 43 75 | 2 4531 3

122Nortn K ...... 25 1 2 11 10 3 19. Carlenton,a, N...... 25 3 1529 1687 14 187 269 5 32 ll 1543 284 20.1.tPleasantsDle NeStrton, N 25...... 25 3 3420 26151'2 10'l124 ~31 12843 71"424 90 21734347 1 7 52| 1 17" 128318432719 33 17

21. Almaer,N ...... 25 5 17912 827 1|||||}1 56 8.taNemlson N.... 255 23 97 126" 132 4 4' 124 52' 3 414' 3 1'2 1 3'1 7 143CheenneiWels, KC.... 25 1617 2684210 898| | 1 20 24. Coliby, K ...... 25 22 4 19 52 6 t|||l4 16 Brwtr~ K .... ~ 25 ~ .. 42 22 1--- I--4 Totaplantsrado..K...... 52 513813103 8 715856617 7 113127 16 18437 20otrattn Cape...... 255 20521 0 13427341534|143|21 2412 7 |1|6 |3 July, 1942 ROLE OF SEEDLINGS 289 grassland. But several of the prairies had some were found in certain dust-covered both fewer species and fewer individuals areas at Quinter, Dighton, and Ness City; than the Fairfield pasture. but seedlings were very scarce to absent Although the true-prairie stations in- elsewhere. Many of these ranges had clude about one-third of the total number, been almost depopulated of native grass, yet 71 per cent of the total seedling popu- and continued drought had prevented even lation was found here. This is in accord these relicts from producing viable seed. with the more favorable conditions for seed production at these stations during DISCUSSION the drought years. Seedlings of sand There is urgent need for restoration of dropseed and blue grama were the most grass on ranges and abandoned dry land uniformly distributed. Dropseed occurred farms. Not only is acute shortage of feed in 19 per cent of the samples and at 19 encountered but also without the protec- of the 22 stations. Blue grama was found tive influence of grass the lands are highly in 24 per cent of the samples and at 18 susceptible to ruin by erosion. Results stations. from germination studies showed that Of the 550 square feet of soil carefully viable seeds of native perennial forage examined, 203 square feet (37 per cent grasses, with rare exception, were present of the area) supported no seedlings. in such moderate numbers (about 26 per Scores of additional areas were examined square foot) in soils of depleted pastures without delimiting them with a hoop. as to be of limited value in restoration of They gave similar results. Numerous the vegetation. sites above, below, and closely adjacent to The abundance of seedlings of sand clumps of cactus, miniature terraces dropseed, both in greenhouse and field formed by deposit, depressions, etc., were studies, are in accord with the records of studied. Bare soil often supported as Weaver and Hansen ('41). They found many seedlings as soil partly covered with 84 per cent of the total grass seedlings debris or vegetation. No relation between which appeared in permanent quadrats at erosion and deposit and position of seed- Lincoln, Nebraska, during 4 years of pro- lings was found. tection of a native pasture from grazing, Seedlings of perennial grasses were were sand dropseed. The seedlings grew rarely more than 2 inches tall. Many, slowly above ground at first but root de- especially sand dropseed, were only 0.25 velopment during the same period was inch in height even in mid June. Seed ample (fig. 14B). This is advantageous depth of blue grama and buffalo grass in establishment of seedlings especially in ranged from the surface to 0.5 inch. The regions subject to intermittent drought. caryopsis was nearly always attached to Sand dropseed, under optimum condi- the base of the seedling. Grass seeds, in- tions, will flower and fruit the first year, cluding buffalo grass, sometimes germi- thus increasing the seed supply. Opti- nated on the soil surface, or rarely in mum conditions, however, occur but droppings of jack rabbits. Sometimes rarely within its midwestern range of dis- seedlings were found lying on the soil tribution. Its emergence throughout surface; others had been partially up- spring, summer, and far into the fall is rooted by erosion, and still others buried also significant. Should the first seed- by deposition of soil. lings succumb to unseasonable weather, Repeated search for seedling grasses others would replace them. This is one was made by Dr. Albertson during June reason for its rather wide use in reseed- and early July of 1941 at most of the re- ing ranges and abandoned lands. It is maining stations in western Kansas. A possible that viable seeds still remained few seedlings were observed in the short- in the samples of soil when the germina- grass habitats at Hays and Phillipsburg; tion studies were concluded. This may 290 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3

FIG. 14. Seedlings of grasses grown out-of-doors during July in compact, fertile silt loam with approximately optimum water content. Scale in inches. A, Bouteloua gracilis; B, Spo- robolus cryptandrus; C, Buchloe dactyloides; D, Panicunn capillare; E, Agropyron smithii; and F, Andropogon furcatiis. All are 24 days old except B, 18 days and E, 20 days. have been true of other species as well, as 50 to 75 short-grass seedlings per since some native grasses germinate better square foot at Hays, Kansas, on June 18, in spring and fall than at other seasons 1938. These did well until the drought (Blake, '35). of midsummer when practically all per- Blue grama grass is one of the best ished. Even those well established were adapted native species for reseeding de- wilting or had become dormant. Bv Au- pleted pastures (Savage, '39). Further- gust 1, myriads of dead seedlings of the more, the distribution and number of short grasses were observed. However, seedlings and their survival in the field about 1 per cent of widely scattered seed- indicate its xeric nature and importance lings of both species survived. Seedlings in natural restoration. The spring and of buffalo grass had as many as 6 to 8 early summer of 1941, however, were ex- tillers each, with leaves about 1.5 inches ceptionally favorable for germination of long. Fewer tillers characterized blue seeds and survival of seedlings. More grama but the grass was taller. Of the often, mortality rate among seedlings is numerous plants excavated, none had high. Albertson in a communication with roots penetrating the soil to a greater the authors stated that he found as many depth than 8 inches. Dead seedlings had July, 1942 ROLE OF SEEDLINGS 291 shorter and more poorly developed roots. withstand drought as well as those of When the supply of soil moisture is ade- most of the preceding species (Weaver, quate, blue grama seedlings tiller freely '30). and flower the first season (Mueller, '41). Sparse branching of roots of seedlings The percentage of germination of buf- of western wheat grass would indicate the falo grass seeds has been found to be necessity for germination of seeds early rather low (Webb, '41). This, together in spring if the plants were to become with poor equipment of the species for established before summer drought (fig. dissemination of seeds, accounts in part 14E). Panicum capillare requires only a for fewer seedlings and their more lim- short growing period for completion of ited distribution. Livestock may be help- its life cycle. It thus escapes drought and ful in disseminating the seeds of buffalo may add a new supply of seeds to the soil grass (Savage, '39), and the seeds may each year. Roots thoroughly ramified be lifted by running water and deposited throughout the soil to depths of 12 to 18 at bases of slopes or in streams (Griffith, inches (fig. 14D), and inflorescences ap- '40). But at no time during this study peared in less than a month. Similar were seedlings found in greater numbers habits of Eragrostis cilianensis contribute in ravines than elsewhere. to its success in early stages of succession. Vegetative propagation by buffalo grass The presence of parent plants and their is very efficient and a seedling may popu- fruitfulness vitally influences the distribu- late a bare space quickly if soil moisture tion and abundance of seedlings. Certain is available (Mueller, '41; Webb, '41). grasses, which were once important con- Tillering begins 10 to 14 days after emer- stituents of the prairie, as little bluestem, gence, and stolons appear when plants are June grass, needle grass, and bluegrass, only a month old. Under very favorable either were missing or occurred but spar- conditions, stolons may elongate at the ingly among the seedlings. Mature plants rate of 0.5 to more than 1.5 centimeters of these species have largely disappeared per day, but when drought comes or if from the midwestern ranges (Weaver and the plants are shaded, elongation is cur- Albertson, '36, '39, '40). tailed. Growth of tops is paralleled by a It is not often that a good seed year vigorous root development (fig. 14C). a favorable for Adventitious roots appear promptly and precedes year germination of penetration often surpasses that of blue of seeds and establishment seedlings. grama at the same age (fig. 14A). Moreover, vitality of seeds collected from Shoots of big bluestem elongate rapidly prairie plants differs f rom season to season and develop much foliage. This is ad- and also according to the degree to which vantageous when seedlings are competing the plants are harmed by grazing (Samp- for light, but when moisture is a limiting son, '14). Climatic conditions at anthesis, factor the ratio of transpiration to absorp- available soil moisture, frosts, etc., are tion by the roots might be great enough important factors. Yield of viable seed to impede further development or even during drought years was practically nil result in death (fig. 14F). Seedlings of in many localities. Yield of caryopses of this species were limited to the true prairie blue grama grass for west central Kansas where the shortage of soil moisture has in 1939 was about 16 per cent of the nor- been somewhat less acute than in the mal (Branson, '.41). Blue grama grass, mixed prairie. Roots of little bluestem buffalo grass, and sand dropseed flowered reach a depth of 6 inches and have at the mixed prairie stations in 1940, but branches of the second order 16 days af- percentage of yield as compared with the ter the plant appears above ground. normal is not known. Grasses which Rapid early tillering greatly increases the flower in summer and autumn were in- transpiring surface. Seedlings do not jured by drought at the prairie stations 292 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3 in 1940. Consequently, little viable seed The period of tillering, before new roots was produced. become well established in the moist layers Owing to local showers, seed may be of soil, is a very critical one for the plant. produced in some parts of the grassland The transpiring area has developed some- every year. During 1937, a year of less what ahead of the root system in contact severe drought, some seed ripened at with deeper soil. Drought at this time Hays, Kansas, and numerous seedlings often causes great loss. Next to germi- were observed in 1938. Most of these nation it is the most critical time in the were growing in the shade of the tufts of life of the plant. For example, hundreds established grasses. They were lighter of seedlings of sand dropseed per square green and more attenuated than those foot occurred in the pastures of eastern growing in the open spaces between the Nebraska in 1940 when the early au- bunches. During seasons of good rain- tumnal rains terminated the late summer fall where foliage development of the drought. Dry weather again occurred established vegetation is good, seedlings and few survived. growing close to the old clumps often The sparseness or absence of seedlings, failed to survive. Conversely, during dry except in the best years such as 1938, has seasons when very little foliage is pro- been repeatedly observed over a wide area duced, seedlings growing under protec- of grassland by Robertson ('39) and tion of old plants are far more apt to R. L. Fowler,4 and by the senior author survive than are those growing in un- since 1934. Savage ('37), in his exten- shaded places between the clumps. Hence sive quadratting, gives similar results. In the presence of ruderals in overgrazed June, 1935, 50 miles southwest of Akron, and dusted pastures is not entirely detri- Colorado, he found an average per quad- mental. Such plants form a protective rat of only one seedling each of blue covering, thus reducing surface soil evapo- grama and buffalo grass in 20 meter ration and curtailing erosion. quadrats. In 38 quadrats near Colby, The probability of the seedlings of such Kansas, the average was 2 and 22 seed- dominant range grasses as blue grama, lings of these grasses, respectively. At buffalo grass, and side-oats grama com- Tribune, Kansas, the average numbers pleting their life cycle by production of from 40 quadrats were 1 and 27 seedlings, seed the first year is remote, unless good but in other heavily grazed ranges 0 to 3. moisture conditions are rather constantly When one considers the chances of de- maintained. A continuous moisture sup- struction of seedlings by drought, tramp- ply rarely occurs in mixed prairie and ling, grazing by cattle, grasshoppers and plains disclimax. Seedlings ordinarily do rodents, disease, winter killing, etc., such not survive unless the adventitious roots numbers are almost inconsequential. become established in moist soil. Devel- On the ranges of the southwestern opment of these is largely dependent upon United States, many important forage soil moisture. The adventitious roots of grasses reproduce from seed only during seedling grasses of the western ranges re- exceptionally favorable years, although quire at least 3 to 5 consecutive days with there may continually be a supply of via- moist surface soil to penetrate sufficiently ble seed in the soil (Glendening, '41). to become established (Cf. Olmsted, '41; This may also prove to be true over large Glendening, '41). But once established portions of the mixed prairie. It has been they grow rapidly and permit the develop- the rule during the present period of great ment of tillers without death by desicca- drought, and wind and water erosion. tion. Chances for survival of seedlings on Mortality by drought of even well es- wind- or water-moved soil are poorer tablished seedlings is often high, and even than on hard, eroded land. The great losses occur following germination. 4Manuscript, 1941. Unpublished. July, 1942 ROLE OF SEEDLINGS 293

transported soil is apt to be moved again, Seedling grasses (6,730) slightly ex- and penetration of rainfall is greatly de- ceeded the number of forbs (6,388). creased in recently deposited soil. More species of grasses grew in samples Seedlings of grasses were found in from true prairie than mixed prairie, but only 63 per cent of the sampled areas. the reverse was true for weedy forbs. Assuming that the perennial grasses had Seedlings of native perennial forbs were been uniformly distributed, the rate of very few. occurrence would have been 4.3 seedlings Three species of native perennial per square foot over the whole area. In grasses, Sporobolus cryptandrus, Boute- the mixed prairie the rate was 2.4 seed- loua gracilis, and Buchloe dactyloides, all lings per square foot. Under the most of high forage value, furnished 73 per favorable conditions, a seedling (except cent of all the grass seedlings. Seedlings those of buffalo grass, which were few) of other perennial grasses were largely would not develop a basal cover of more confined to prairie stations and were rel- than one square inch in a single summer. atively unimportant. Providing all of the seedlings in mixed Numbers and kinds of seedling grasses prairie lived and grew to this size, they were determined in June in each of 25 would furnish a new cover for somewhat square-foot areas in each range or prairie. less than 2 per cent of the area. Thus, Extensive soil sampling and study of nu- despite the favorable year, the discrepancy merous rainfall records showed that an between the greenhouse production of na- almost continuous supply of moisture had tive perennial forage grasses of about 26 been available to promote germination and per square foot and those found in the establishment of seedlings. field (about 4.3 per unit area) is very The rank growth of abundant weeds great. and much depleted cover of perennial It is indeed fortunate that abundant forage grasses were studied in relation to rainfall over most of this grassland in seedling development. 1941 has promoted excellent seed produc- Twenty-four species of seedling grasses tion. If this is followed by another year were found in true prairie but only 7 in conducive to germination, seedling devel- mixed prairie. Mixed prairie included opment, and another crop of seeds, re- about two-thirds of the places examined vegetation will be greatly promoted. but only 29 per cent of the total seedlings Complete restoration of the former com- were found here. position and density of the vegetation will Of 550 square feet of soil on which require a long period of time and most seedling grasses were counted, 37 per cent judicious range management. supported none. Seedling grasses were especially rare in drought-stricken and SUMMARY dust-covered ranges of western Kansas. Square-foot samples of surface soil Seedlings of Sporobolus cryptandrus were collected from 49 drought-damaged and Bouteloua gracilis were the most ranges and prairies in Nebraska, Kansas, widely distributed and most abundant of and Colorado, and viable seeds germi- those of perennial grasses. nated. Seedlings grew at the average rate Viable seeds of native perennial forage of 67 per sample. Forty species of forbs grasses, with rare exception, were pres- occurred, of which more than 96 per cent ent in such small numbers (26 per square were annual weeds, mostly Arnaranthus foot) as to be of limited value, when spp. and Salsola pestifer. There were 26 seedling hazards are considered, in resto- species of grass seedlings of which 20 per ration of the vegetation. cent were ruderals, mostly Eragrostis Average distribution of perennial grass cilianensis, Hordeum pusillum, and Pani- seedlings on the ranges and prairies was cum capillare. 4.3 per square foot; in mixed prairie 294 J. E. WEAVER AND IRENE M. MUELLER Ecology, Vol. 23, No. 3 alone, 2.4 per unit area. Even if all seed- Mueller, I. M. 1941. An experimental study lings (exclusive of the stoloniferous buf- of rhizomes of certain prairie plants. Ecological Monog., 11: 165-188. falo grass) had survived and made a Olmsted, C. E. 1941. Growth and develop- maximum growth in mixed prairie, they ment in range grasses. I. Early develop- would have increased the cover less than ment of Bouteloua curtipendula in relation 2 per cent. to water supply. Bot. Gaz., 102: 499-519. Several years of good seed production Riegel, A. 1941. Life history and habits of blue grama. Trans. Kans. Acad. Sci., 44: and development of seedlings into mature 76-85. grasses are necessary for the restoration Robertson, J. H. 1939. A quantitative study of midwestern ranges. of true-prairie vegetation after three years of extreme drought. Ecological Monog., 9: 432-492. LITERATURE CITED Sampson, A. W. 1914. Natural revegeta- Albertson, F. W. 1941. Studies on the tion of range lands based upon growth prairies of southwestern Kansas. Trans. requirements and life history of the vege- Kans. Acad. Sci., 44: 48-57. tation. Jour. Agr. Res., 3: 93-148. Blake, A. K. 1935. Viability and germina- Savage, D. A. 1934. Methods of reestab- tion of seeds and early life history of lishing buffalo grass on cultivated land in prairie plants. Ecological Monog., 5: the Great Plains. U. S. Dept. Agr., Circ. 405-460. 328. Branson, L. R. 1941. An analysis of seed . 1937. Drought survival of native production of native grasses during the grass species in the central and southern great drought. Trans. Kans. Acad. Sci., Great Plains, 1935. U. S. Dept. Agr., 44: 116-127. Tech. Bull. 549. Clements, F. E., and J. E. Weaver. 1924. . 1939. Grass culture and range im- Experimental vegetation. Carnegie Inst. provement in the central and southern Wash. Pub. 355. Great Plains. U. S. Dept. Agr., Circ. Copple, R. F., and A. E. Aldous. 1932. The 491. identification of certain native and natu- Weaver, J. E. 1930. Underground plant ralized grasses by their vegetative char- development in relation to grazing. Ecol- acters. Kans. Agr. Exp. Sta., Tech. Bull. ogy, 11: 543-557. 32. Weaver, J. E., and F. W. Albertson. 1936. Glendening, G. E. 1941. Development of Effects of the great drought on the seedlings of Heteropogon contortuis as re- prairies of Iowa, Nebraska, and Kansas. lated to soil moisture and competition. Ecology, 17: 567-639. Bot. Gaz., 102: 684-698. . 1939. Major changes in grassland as Griffith, D. E. 1940. The collection and a result of continued drought. Bot. Gaz., processing of buffalo-grass seed. Soil 100: 576-591. Cons., 6: 132-133. 1940. Deterioration of midwestern Harrington, H. D. 1938. Key to some ranges. Ecology 21: 216-236. Colorado grasses. Mimeographed leaf- Weaver, J. E., and W. W. Hansen. 1941. let. Colorado State College, Fort Col- Regeneration of native midwestern pas- lins, Colo. tures under protection. Nebr. Cons. Bull. Keim, F. D., et al. 1932. The identification 23. of the more important hay grasses of Ne- Webb, J. J., Jr. 1941. Life history and braska by their vegetative characters. habits of buffalo grass. Trans. Kans. Nebr. Agr. Exp. Sta., Res. Bull. 65. Acad. Sci., 44: 58-75.