BACTERIOLOGICAL REVIEWS, Dec. 1969, p. 476-504 Vol. 33, No. 4 Copyright @ 1969 American Society for Microbiology Printed Ln U.S.A. Thermophilic Blue-Green Algae and the Thermal Environment RICHARD W. CASTENHOLZ Department of Biology, University of Oregon, Eugene, Oregon 97403

INTRODUCTION...... 476 DISTRIBUTION OF THERMAL WATERS...... 481 SOLUTES OF THERMAL WATERS ...... 482 DISTRIBUTION OF SPECD DS ...... 486 Upper and Lower Temperature LLmts...... 486 Clas tion and Geographical Distribution ...... 487 Problems of Survival and Transport...... 488 STUDIES OF NATURAL POPULATIONS ...... 489 Mat Formation and Stability...... 489 Movements of Filaments and Mats ...... 491 Measurements of Photosynthesis and Growth ...... 491 CULTIVATION OF THERMOPHILIC CYANOPHYTES ...... 493 Medium and Nutrition ...... 493 Isolation and Maintenance...... 494 Rates of Growth, Photosynthesis, and Respiration in Culture...... 495 RESPONSES TO TEMPERATURE AND LIGHT INTENSITY...... 497 Optimal Temperature and Light Intensity...... 497 Effects of Light and Temperature on Pigmentation ...... 498 Growth and Survival at Suboptimal Temperatures ...... 499 LITERATURE CITED...... 500

INTRODUCTION most striking effects, the promotion of blue-green The blue-green algae (cyanophyta) are con- algal growth (164). It is probable that habitats sidered to be thermophilic (in this review) when suitable for thermophilic organisms are going to part or all of their optimal growth temperature increase substantially. range is above 45 C. Different definitions are used Most thermal habitats are aquatic, and the for thermophily in bacteria, fungi, and animals. source of heat is telluric for nearly all of these. A maximal growth rate at temperatures over 45 C Nevertheless, insolation can raise the temperature is mainly a characteristic of procaryotic or- sufficiently in a few situations and the self-heating ganisms. Only a few species of eucaryotic protists of organic materials (thermogenesis) may bring or animals tolerate temperatures above this localized temperature to the point of ignition. (Table 1). In the range between 50 and 60 C, there Hot springs and their drainways provide the are a few fungi and the eucaryotic alga of acid most abundant aquatic habitats for thermophilic waters, Cyanidium caldarium. Photosynthetic blue-green algae. It will be these natural environ- blue-green algae are known to grow at constant ments and these blue-green algae that are dis- temperatures as high as 73 to 74 C (29), and cussed through the remainder of this paper. nonphotosynthetic bacteria as high as 95 C (26, However, the other (rarer) thermal habitats Table 1). Even many species of blue-green algae require a brief description, even though little is of nonthermal habitats have higher temperature known of their organisms in situ. Temperatures in optima than the eucaryotic algae of the same excess of 50 C may be attained in a few truly waters (79). Because of this, blue-green algae may aquatic situations solely from insolation. The be enriched for by incubating samples in light and monimolimnion (bottom waters) of a few small nonselective mineral medium at 35 C (11). meromictic saline ponds and lakes may be heated Blue-green algae are becoming more con- (to >50 C) during the seasons of high light spicuous in this age of increasing environmental intensity and retain fairly high temperatures pollution. In nutrient-enriched waters, blooms of throughout the year, since circulation of the bot- planktonic blue-green algae are more frequent, tom water is completely lacking (15, 95). Con- denser, and longer lasting (80). Similarly, thermal siderably higher temperatures (90 to 95 C) were pollution from the water coolant of power plants reached in artificial solar ponds at the Negev (nuclear and conventional) has, as one of its Institute in Israel (J. Schechter, personal com- 476 VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 477 TABLE 1. Organisms that live in hot springs at isolations of these organisms can easily be made temperatures above 45 Ca from hot springs (134) and a variety of non- Organism Temp thermal habitats such as soil, sand, air, seawater, and snow (5). Filamentous and unicellular bacteria Although the source waters of hot springs are [mainly heterotrophic ? (26, 29, 32) ] ...... 95 C usually quite low in dissolved organic compounds, Acidophilic thiobacilli (34) ...... about 60 C there are obviously a number of heterotrophic Photosynthetic (purple sulfur) bac- bacteria associated with the photosynthetic blue- teria (unpublished data) ...... 57 to 60 C green algae on which they probably depend for Photosynthetic blue-green algae (31, 38, carbon-energy sources. Some of these are fila- 146) ...... 74 C mentous flexibacteria; others are nonsporeform- Cyanidium [(acidophilic) (16, 34)]. .. 56 to 57 C Fungi [(?) in hot springs (58)] ...... 60 C ing filaments or rods of other types (42). How- Diatoms (120, 169; R. P. Sheridan, personal ever, in many hot springs in Yellowstone (26) communication) ...... 50 C and Oregon (unpublished data) there are bacteria Green algae (120, 139, 169) ...... 48 C growing above 90 C where no endogenous photo- Ciliata (45) ...... 50 C synthetic organisms are present to support Rotifera (45) ...... 45 C heterotrophic growth. Either the rapid flow of Ostracoda [crustacea (45; Castenholz, spring gyater (containing a minute quantity of unpublished data)] ...... 49 to 50 C organic matter) is sufficient to sustain growth or Acarina [arachnida (45, 46)] ...... 50 to 51 C Diptera [larvae (28, 45, 46, 175) ] ...... 50 C these bacteria are autotrophs of some type. Coleoptera [larvae (45)] ...... 45 C Blue-green algae are particularly concentrated in hot-spring waters with a pH of over 6 where a Approximate highest constant temperatures they form conspicuous and often unialgal matlike at which they occur or at which growth has been covers over submerged substrates. Since there is demonstrated are indicated. Selected references in many hot springs a surface effluent with a on temperature limits are given for each group of thermal gradient ranging from supraoptimal to organisms. Except for Cyanidium and certain bacteria and fungi, all of the organisms occur ambient air temperature, specific differences in mainly in waters ofpH above 6. growth temperature optima may result in distinct species bands covering different portions of the gradient (Fig. 1). Since the component organisms munication). In these, meromixis was established differ in their amounts of chlorophyll-a, bilipro- by artificially increasing the salt concentration teins, and carotenoid pigments, these bands may towards the bottom. Small freshwater or saline be quite different in color, but ranging from a dark pools in warm desert or tropical island environ- brown to a yellow or rich green or blue-green. ments sometimes exceed 40 C (B. Whitton, The orange or flesh color of many hot spring mats personal communication), although air tempera- is often caused by compact masses of hetero- ture is also important in such circumstances. The trophic, carotenoid-containing, filamentous bac- origin of heat in the deep, hot brines of the Red teria. Although these generally form gelatinous Sea is telluric, and only bacterial life has been layers (sometimes a few centimeters thick) under- reported to date (174). Blue-green algae are the neath a top cover of photosynthetic blue-green principal inhabitants of alternately moist and dry algae, they are sometimes exposed over wide cliffs in most mountainous regions of the world. areas (Fig. 1, 2). In some springs, pink to red The conspicuous dark streaks of blue-green algae layers of photosynthetic purple sulfur bacteria may be heated by sun and air to temperatures of occur directly under the algal cover. over 40 C (109), but it is not known that Because of the shallowness and the clarity of growth occurs during such periods. most thermal waters, and the exposure of many Hot springs and most solar-heated aquatic hot springs to high light intensities, various types environments are composed of mineral waters, of "sun adaptation" have occurred in many whereas the environment in self-heating piles of thermophilic organisms. Laboratory and field vegetation is richly organic. Thermophilic fungi, studies of this have added an interesting dimen- sporeforming bacteria, and actinomycetes appear sion to studies of adaptation to high temperature. to be the main agents of thermogenesis in ac- In addition, the organisms of many hot springs cumulations of hay, compost, peat, and manure. have adapted to high salinity or to high concentra- A few of the fungi are active between 50 and tions of certain ions. Being partly of magmatic 60 C, whereas some of the bacteria are active to origin, most thermal waters differ considerably about 70 C (5, 58, 77). Although self-heating vege- in their chemistry from surface waters of lakes tation may be one of the main habitats for the and streams (Table 2; 127, 183). Most alkaline growth of thermophilic spore-forming bacteria, hot springs contain between 1,000 and 2,000 478 CASTENHOLZ BACTERIOL. REV. mg of total dissolved solids (TDS) per liter; some several other elements (Table 2). Although the have salinities higher than that of seawater. Even gross chemistry of a large number of hot springs those of lower salinity may be notably enriched is known and there is some information on the in S (reduced), As, F, Mn, Fe, Al, Li, Si, or in nutrition of thermophilic cyanophytes, few

Fir. 1. Hunter's Hot Springs, Lakeview, Ore.: a portion of the main thermal stream. The arrow indicates the direction offlow. The distance between the no. I and no. 3 markers is approximately 0.5 m. (1) Green-colored Synechococcus cover, >54 C. (2) Dark brown 0. terebriformis cover, <54 C. (3) Pink-orange exposed undermar offilamentous bacteria, < about 45 C, plus several cyanophytes. The water depth rangesfrom afew millimeters to about 5 cm in midchannel. 30 March 1965. VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 479

FIG. 2. Hunter's Hot Springs: a closer view of a mat, with a portion cut away. (1) Green-colored top cover of Synechococcus. (2) Brown-colored 0. terebriformis top cover in various contraction patterns and overriding the Synechococcus. (3) The undermat of pink-orange filamentous bacteria. Total mat thickness is about 2 cm. The arrow indicates the many separable layers of the undermat. The 5-cm scale marker is on the sandy substrate. Slow-moving water here was 1 cm or less deep over the top cover and about 53 C. 29 May 1967. attempts have been made to correlate field and years ago) resemble living members of the blue- laboratory data. green algae (151). Most of the fossil-rich Pre- One of the most interesting ecological features cambrian material occurs as chert or as stromato- of most hot springs is the great constancy of lites which could have originated in aquatic temperature and chemistry of the water at the thermal environments similar to those of today. source, generally unequaled by other aquatic In particular, the description by Schopf (151) of systems. In practice, this means that a thermo- the probable conditions of deposition of the late philic alga has its optimal range of temperature Precambrian cyanophyte-rich material sounds available in the thermal gradient at any season. very similar to the manner in which blue-green Similarly, any chemical limitations that exist algae are embedded by siliceous sinter in con- exert their influence year round. This is quite temporary hot springs of high or moderate unlike the situation in typical streams and lakes salinity. The question of whether the adaptations where great seasonal fluctuations of temperature of modem thermophiles to high temperature and nutrient content may occur. In addition, the occurred in Precambrian or more recent times is relatively few species of each trophic level greatly unanswerable at present (see reference 5). simplify studies of species interactions and of The physical-chemical bases of the stability of productivity. compounds and cellular structures at elevated It is generally assumed that blue-green algae temperature are still poorly understood. Most of evolved in the early Precambrian and were re- the work has been with the thermophilic spore- sponsible for the first significant increase in at- forming bacteria, primarily Bacillus stearother- mospheric oxygen. Microfossils of cyanophyte- mophilus (e.g., 3, 12, 18, 20,25, 29, 34, 48, 50, 56, like filaments have recently been found in early, 72, 86). It may be fallacious to extrapolate these middle, and late Precambrian strata (171). results in toto to the blue-green algae which are Fossils of the late Precambrian (about 1 billion photosynthetic and from a different chemical 480 CASTENHOLZ BACTERIOL. REV.

"0 %n 0 -N 0

V ~00 0oN0 C Joo O J0 ' ON 4 a " en en NO f I I III , _ r- 0 Nr 0 1- s-U 0 0

-0

4' 0 cdW

co - fDO0-Nr)0 n - I0O - -IO10000c 0008 0° 0 I 008o - 0 v _0 0

...... 10 ...... I. 00\00 0 -0 0 000O000 0 0 00 en 0 0

_ n "wN t . V . 00 00

N en ,4Cle.I0 4 (D I -e 1(4,N, lo 10 liS ~6ci- 1I 1I 1I 1I en I"U

o 0- z bo~o-0ooroIx o Nr~o~00oafoena'I'o 000oo~o 00oaoo00 o>Nro - UV0 C! ! X.o A X - ° °

F-oI onl=(O Iooo I V I I I )0 =Z E 4i-, L. 00 Q 4- Q -C

1. '4 . 0 00 *10 09 -08 08 m~t O~totg8Oi m _ ..- [1 -oo.t 4NOEci N.- 001t-W cq 0 000000cIc ooo-o0 n-N - III .II IIn 4!o&J(U', .%0 00N (6~~0~O~c; cIr; i N 1. _ In 0 0 N N o 0~~ ~ ~ ~~~- ,0 _ O0..8ma00O O r-NoREa ZLa)YUu S t 1 ^ 1 III IIIIIII bN O00;; 3W;o en~O Io0N ,-4r 0 E E

NO 00 %O 00~~~~~~~~~~~~~~~0 0ooIn lz 00C N0' 01 t ooe- 0%ct* ;>, I o 2 ...... ci.0N00 Ne'a0 C)a 4:.61~~ ~ ~ ~ ~ ~ In0 .1 I 0f cls~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~r 8N l. ci~on0N t-0_000 0N000 0) 0* W)I00% n ( ri C'0 N a) .) U2 b e t oei N~~~~~~~~~0 na

Ls- . O 0 i, 00 tnenS8 - 0,-0000 0 -'ra-000 0 0v' 8 RnCl"A-00rIOr-.4 'IO ton~'t0-NCI 0 -4 ON - en

co . . : : : -.o . co 0 tkoo u v:~~~~~~~~~~~~~~XZz . < . n O. ,_; ze C 6czu) * * z o0 z u z 3 < L~1= W4 q~ F4 VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 481 environment. It has been suggested (29) that the (3, 17, 176). Flow rate may be less constant, photosynthetic apparatus itself is the most particularly in geyser areas where the well-known temperature-sensitive component of the blue- surges of water cause great changes in drainway green algae with the highest temperature toler- flow and temperature over short periods of time. ances (i.e., 73 to 74 C). On the basis of very little Springs in which the water discharges at a information, thermophilic blue-green algae appear temperature of over 45 C occur in every continent to have the same functionally defined and struc- except Antarctica (Fig. 3). Most of these spring turally defined systems as other blue-green algae areas have been mapped, listed, and referenced (8, 9, 69, 121, 132). Among thermophilic types, by Waring (181). Concentrated thermal activity only Synechococcus lividus and Mastigocladus is usually associated with Tertiary or Quaternary laminosus have been described at the ultrastruc- volcanism, where magma may still lie closer to tural level (69, 132), and even the gross aspects of the surface (Fig. 3). Many groups of springs, photosynthesis, respiration, and other metabolic however, are associated solely with the deep processes at high temperature are undescribed penetration of water in fault zones. The greatest in most species (52, 133). concentrations of thermal springs occur in the The isolation of thermophilic variants of meso- Yellowstone Plateau of North America, the philic bacteria capable of growth at 55 C has been North Island of New Zealand, Iceland, and consistently successful with several spore-forming Japan. Large numbers of hot springs are also types (5). The ability to grow at 55 C has also widely scattered over most of the United States been genetically transformed to nonthermophilic west of the Great Plains, the line of the Andes, bacteria by mesophilic forms, indicating that the Italy, Algeria-Tunisia, Greece, Turkey, portions original thermolability may have resided in a of central Africa, India, central Asia, Indonesia, single system or component (21, 136). In view of Melanesia, the Philippines, and Kamchatka (Fig. this, it seems possible at least that all cell systems 3). Australia has few hot springs and the whole of mesophilic procaryotes such as blue-green of northern Europe above 52° lat is devoid of algae may be capable ofoperating at temperatures thermal waters above 40 C. A few hot springs lie of about 55 C with only minor genetic alterations. on the Greenland coast 300 to 400 miles to the Perhaps 55 to 60 C represents a borderline, above west of Iceland, but all of North America east which cells are required to use very different and of the Rocky Mountains lacks thermal water, perhaps more cumbersome methods of assuring except for the few springs in Virginia, North thermostability, methods which may prevent the Carolina, and Arkansas (181). operation of metabolic systems at lower tem- Although there is a great variety of chemical peratures (22). types, there seem to be thermal waters on every continent that are essentially identical to some DISTRIBUTION OF THERMAL WATERS on almost every other continent. Any area of The water of hot springs may discharge at intense thermal and volcanic activity usually temperatures somewhat higher than the common has the greatest variety of spring types. The major boiling temperature, although superheated waters areas of Yellowstone, Iceland, and New Zealand are uncommon except in volcanic regions (17). have probably been active continuously since the Superheated steam issuing from vents may have early Pleistocene. It is likely that throughout the considerably higher temperatures (2, 17). A spring biological history of the earth thermal water may discharge into a basin, thereby forming a has existed, although not in the same areas. pool or lake [limnotherm (179)]. Some of this Terrestrial volcanic belts are not connected by type have no surface effluents. The pool itself underground water systems (13), although this (with or without an overflow) may be of a uni- has been suggested. form temperature if the water is well mixed by a Exposure to extremely high light intensity is bottom source of gas or water. If less turbulent, almost universally characteristic of hot spring water along the edges may cool slightly. Other environments. Most of the extensive thermal pools have stream outlets. Whether thermal areas are devoid of higher vegetation, or nearly streams originate from pools or directly from the so. In addition, some of the best known areas earth [rheotherm (179)], they will display an entire are at high elevation or in regions of little over- temperature gradient from source temperature to cast. Since the water of most thermal pools is close to ambient atmospheric temperature unless clear and the effluent streams are usually shallow, truncated by the flow entering another body of there is little radiation extinction. As a result, a water or disappearing into the ground. Although large number of thermophiles must adapt to high many springs have quite constant temperatures light-level situations or else physically avoid the at their sources over many years, others vary light. In Yellowstone, (elevation 6,000 to 8,000 considerably in the course of years, days, or hours ft) exposures to 0.6 cal per cm2 per min of radia- 482 CASTENHOLZ BACTERIOL. REV.

FIG. 3. Global distribution of thermal areas with springs discharging at temperatures over 45 C (0) and of contemporary or Recent volcanism (A). The regions with particularly high densities ofhot springs are indicated by solid circles and arrows. Most of the information has been obtained from Waring (181). Capital letters refer to selected hot-spring areas where floristic studies of the component blue-green algae have been made. Major refer- ences are: A, Oregon (145); B, Yellowstone Park (59, 139); C, Iceland (23, 27, 54, 139, 153); D, France (73, 139); E, Hungary, Czechoslovakia, and Yugoslavia (57, 115, 116, 139); F, Greece and Israel (14, 62, 139); G, India (173); H, Indonesia (91); I, New Zealand (139); and J, Japan (71, 187). The reference to Nash (139) is included for several regions because ofhis comprehensive review of historical colections (before 1923) from European and North American hot springs. tion in the visible range (400 to 700 nm) are probably varying with the increase or decrease common at midday in summer, whereas similar of precipitation in the watershed. Thus, the intensities occur in the hot-spring areas of eastern ratios of certain elements and the total salinity Oregon (altitude 1,500 to 4,800 ft). Such high may change considerably in a short period (176). intensities, with spectral distributions similar to Nevertheless, the great majority of thermal overcast or sun-plus-sky radiation, have been springs appear to have remarkably constant approached in the laboratory with quartz iodine chemical composition and temperature at the vapor lamps and appropriate liquid filters (110). point of emergence. Springs of somewhat lower temperatures SOLUTES OF THERMAL WATERS (<80 C), in which the waters are wholly of The solute concentration of hot springs may recent meteoric origin, generally have lower vary greatly, even within a local area (2, 17, 61, salinities, often between 200 and 400 mg/liter 181, 183). Thermal water with less than 500 mg (183; Table 2). Highly saline hot springs (>20,000 of TDS per liter is uncommon. The median mg of TDS per liter) occur in widely scattered concentration for the earth's thermal springs is regions, associated with deep sediments or probably only slightly less than 2,000 mg of volcanism [e.g., Utah, Israel, Turkey, Greece, TDS per liter. This is considerably higher than Japan, Philippines, Indonesia (181, 183; Table the typical value (less than 150 mg/liter) for 2)]. In a highly saline spring on Reykjanes (SW surface waters of lakes and streams (127; Table Iceland), with a salinity of over 45,000 mg of 2). Hot subsurface waters are rich in solutes TDS per liter, no algae were found (T. D. Brock, because of their greater leaching and carrying personal communication). The pH of this spring capacity. It is generally assumed that the mineral is below 7, and the water may contain poisonous content of a hot spring remains quite constant. concentrations of some elements (183). Blue- However, erratic or seasonal changes may occur, green algae are well known in highly saline lakes, VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 483 and many also thrive in certain saline hot springs. high ( > 120 mg/liter). Most volcanic acid springs The Tiberias Hot Springs of Israel, containing (e.g., North America, Japan, Indonesia, New over 30,000 mg of TDS per liter with a pH from Zealand, Italy, Iceland) are inhabited by the 6.2 to 7.0 (183), have an abundant blue-green photoautotrophic eucaryote, Cyanidium calda- flora which includes about 35 species, according rium in the thermal range below 56 to 60 C (34, to Dor (62). In Greece, hot springs with over 54). Highly saline acid springs are also common, 15,000 mg of NaCl per liter also have several but I am not aware of any reports of photo- blue-green species (14). autotrophs in such waters (Table 2). Salinities of It is difficult to estimate to what extent the from 35,000 to 100,000 mg/liter are known in water of thermal springs is of magmatic origin. In western parts of the Pacific circle of contemporary general, very hot waters with deep subterranean volcanism (183). sources are partly derived from magmatic steam, (iii) Calcareous, travertine-depositing springs although an ultimate meteoric source probably are widely scattered, and are probably in contact accounts for the major portion. Magmatic con- with limestone deposits (183). They are generally tributions are more significant in regions of near neutrality and have higher concentrations Tertiary or Quaternary volcanism, where the of Ca, Mg, and HCO3 (relative to sodium and magma is still relatively close to the surface. chloride) than neutral or alkaline springs of the The water of a few springs, both thermal and volcanic type (Table 2). Hydrogen sulfide may be nonthermal, is of a connate origin (i.e., trapped high in some springs of this type. They precipi- in sediment at the time of its deposition). Since tate large amounts of calcite or aragonite trav- the origins of thermal waters differ, so do the ertine. This material is also deposited by the relative concentrations of numerous ions; there common volcanic-Na, Cl-HCO3 springs, but not is no simple method of classification. in such large amounts. The water is mostly, if not Somewhat condensing a general classification entirely, of meteoric origin. All travertine-deposit- of subsurface waters used by White, Hem, and ing springs are supersaturated with CO2. As a Waring (183), we may list the common types of result of the decrease in pressure upon surfacing, thermal waters (based on origin and chemistry) CO2 is evolved and CaCO3 is precipitated. as: (i) volcanic-Na, Cl-HCO3; (ii) volcanic-acid, (iv) Meteoric, low-salinity (<500 mg of TDS S04; (iii) calcareous, travertine-depositing; (iv) per liter) thermal waters are characterized by meteoric-low salinity; (v) brine-Na-Ca, Cl. their association with diastrophism rather than Examples of the chemical compositions of these volcanism. Nitrogen is usually the major gas. general categories are given in Table 2. Sodium or Ca is the dominant cation and Cl is (i) Volcanic-Na, Cl-HCO3 springs are neutral usually low compared to HCO3 or even S04 to highly alkaline waters which occur near Ter- (Table 2). Temperatures are often lower than tiary or more recent volcanic activity and are those of volcanic springs, although the circulation partly derived from magmatic steam. The princi- depth may reach thousands of feet (183). pal solutes are sodium, chloride, and bicarbonate (v) Thermal brines (Na-Ca, Cl), resembling (or carbonate), and silicate (Table 2). The chlo- oil-field brines, are not common but occur in ride may be largely magmatic. Salinities are many geographical regions. Salinites range from generally between 1,000 and 3,000 mg/liter in below to well above that of sea water, and the geyser areas (i.e., high temperatures near the pH is usually between 6 and 7 (Table 2). These earth's surface), but may be higher elsewhere waters are thought to be connate (183). High (183). Almost all the hot springs of the Yellow- concentrations of methane are common. stone Plateau (>3,000) have salinities in this Hot acid sulfur springs and acid brines provide range (2). They usually also contain over 200 mg such restrictive conditions that blue-green algae of silica per liter, in which case opaline (siliceous) are either absent or very rare. However, most sinter may be deposited. The sulfate concentration the neutral- may be high but it varies with the spring (Table hot springs, including common 2). This category of spring comprises the bulk of alkaline types of the volcanic areas, are greatly the alkaline hot springs of the earth. enriched with many ions rare in the surface waters (ii) Volcanic-acid S04 waters are closely of lakes and streams (Table 2). Some of these associated with volcanic activity and surface elements are micronutrients, but are present at fumaroles (solfataras). Sulfuric acid, resulting in concentrations that are probably toxic to many pH values between 1 and 4, originated mainly microorganisms. Some are lost from solution from the oxidation of the abundant sulfides in soon after surfacing, but this may depend on the the subterranean water. Hydrochloric acid (mag- temperature, pH, and oxygen tension. Uzamasa matic chloride) may contribute to the acidity in (176) reported the following concentrations as some waters (Table 2). Silica is generally quite extremes from a variety of volcanic hot springs 484 CASTENHOLZ BACTERIOL. REV. in Japan (mg/liter): Al, 1,000; As, 5.1; Co, 2.19; abundant than in nonthermal surface waters Cu, 68; F, 16; Fe, 1,000; Pb, 2.6; Ni, 9.38; Mn, (176) 278; and Zn, 2.0. The greatest enrichment of the The macronutrients which are most often metallic ions is in the more acid waters (Table 2). limiting in lake waters are inorganic phosphate The mean copper level was about 0.098 mg/liter and combined nitrogen. Thermal waters, al- in the acid springs of Hokkaido, whereas concen- though of comparatively high salinity, did not trations between 0.020 and 0.030 mg/liter oc- acquire their salts in the same evaporative manner cuffed in the neutral-alkaline springs (176). as surface waters of closed basins which are Manganese is also higher in the acid springs, but usually enriched in both, but particularly in P04. waters with a pH of up to 8.0 may still contain From the relatively few analyses available, it large quantities. The mean for Mn-containing appears that NO3-N may be very low or lacking springs in Japan is 2.3 mg/liter, an extremely entirely in many hot spring sources (30, 52, 183). high level compared to lake or stream water. However, even low levels may not limit microbial Dark oxides of Mn and Fe are characteristic growth if the flow rate is sufficiently high. Springs deposits in the drainways of many Na, Cl-HCO3 which do not have detectable amounts of NO3 springs. Aluminum and Fe are commonly at at their sources may acquire higher levels down- concentrations over 10 mg/liter in volcanic-acid stream, either from surface drainage or as a result waters, and extremes of over 1,000 and 10,000 of N-fixation. In Hunter's Hot Springs of eastern mg/liter, respectively, have been reported (183; Oregon, NO3 was undetectable at one source, Table 2). Arsenic may be high in acid or alkaline but ranged from 0.042 to 0.142 mg of N03-N springs, and values over 2.0 mg/liter are common per liter 20 m downstream (52). Brock also found in Yellowstone (2) and in other alkaline hot an increase in nitrate and ammonia in a down- springs (183). In 174 neutral and alkaline springs stream direction (30). NO3 ion was found in 15 % in Japan, the mean concentration was 0.28 mg/ of the Japanese springs, presumably at the sources liter. About twice as much occurred in 16 acid (176). In those, the mean level was about 0.2 springs (176). Fluoride is generally higher in mg/liter. Much higher values may occur in more mildly alkaline waters. For Japanese springs, saline thermal waters (183). Among the blue- the concentration was generally between 1.0 green algae there is increasing evidence that and 2.0 mg/liter. Yellowstone springs are noted N-fixation is restricted to heterocyst-containing for high fluoride concentrations, commonly species, possibly to the heterocyst itself (76), but between 15.0 and 20.0 mg/liter (2; Table 2). apparently exceptions may occur (186). Among Bromide concentrations of over 1.0 mg/liter blue-green algae which grow above 54 C, only are most common in more saline springs of either M. laminosus has heterocysts, and N-fixation the volcanic or connate types (Table 2). Sulfides has been demonstrated in axenic cultures and may be common in springs of diverse types. cell-free preparations (78, 152). Steward has also Tolerances of blue-green algae to most of the shown in situ N-fixation in hot spring drainways above solutes have not been established (see dominated by Mastigocladus (168). reference 94). The high tolerances of some blue- Ammonium-N is generally more abundant green algae are evident from their growth in hot than NO3 or NO2, at least at the spring source, springs containing large amounts of some of these in both alkaline and acid types, but particularly elements. Some of the obvious differences in flora in the latter (183; Table 2). Ammonium ion was and productivity among hot springs may result found in less than one-half of 860 hot springs in part from the different concentrations of analyzed in Japan; in these it averaged about 1.6 "minor" elements. Perhaps the frequently con- mg/liter-considerably higher than NO3 (176). spicuous absence of eucaryotic algae below about The depressed upper temperature boundary for 25 to 30 C in drainways of many hot springs Synechococcus in some springs might be attributa- (particularly in volcanic types) is explicable by ble to source waters very low in all forms of their relatively high sensitivity to certain ions. combined nitrogen, although such correlations Practically all analyses of thermal springs have have not been made. been made by geochemists or hydrologists. Inorganic phosphate analyses of thermal water Consequently, analyses of biologically important sources often show values much higher than elements have been incomplete. The elements those found in surface fresh waters (30, 127, 170, known to be micronutrients for blue-green algae 183; Table 2). Nevertheless, P04 was detectable (e.g., Cu, Co, Zn, Mn, Fe, B, Mo) are often in only 37% of the Japanese springs; in these it omitted from the analyses unless their presence averaged about 6.5 mg/liter-a very high value in large quantities is suspected. Except possibly when compared to surface waters (176). Mush- for Mo and Co, these elements are usually more room Spring (Yellowstone), Brock's main re- VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 485 search spring, contained 2.7 mg of P04/liter- TABLE 3.-Continued also high (30). In some springs, phosphate values may increase downstream if drainage or seepage P. purpurascens (Kutz.) Gom. (un- published data) ...... 46-47 C from the shore occurs. In Hunter's Hot Springs, a B, C, D, I, J concentration of <0.040 mg of P04-P/liter at Symploca thermalis Kutz. one source increased to over 0.100 mg of P04- var. (unpublished data) .45-47 C or (?>50 C) P/liter (i.e., 0.300 mg of PG4 ion/liter) 20 m (A?), C, D, F, J downstream (52). It is probably safe to state that most hot springs Order: Nostocales Family: Rivulariaceae Calothrix sp.[incl. C. thermalis (Schwabe) Hansg. (145)]...... (52-54 C) TABLE 3. Blue green algae of thermal springs" A, B, C, E, J Order: Chroococcales Family: Chroococcaceae Order: Stigonematales Family: Stigonemataceae Synechococcus lividus Copeland var. (31, Mastigocladus laminosus (Ag.) Cohn 38, 146) ...... 74 C (54, 154) ...... 63-64 C A, B, (I?), J Second type (54) ...... 57 C "Type IV" (146) ...... 72 C A, B, C, D, E, F, G, H, I, J "Type III" (146) ...... 67 C a "Type II" (146) ...... 62 C Selected list of species that are common in at "Type I" (146) ...... 58 C least some thermal areas. The upper limit without parentheses refers to the maximum constant tem- Synechocystis elongatus Naeg. var. (14, perature at which growth has been demonstrated 91) ...... (66-70C) (in the field or in culture); the upper limit with A, B, E, F, G, H, (I?), J parentheses is based on certain observations of occurrence only. The references apply mainly to S. minervae Copeland var. (14, 145) ..... (60 C) temperature limits. The distribution of species A, B, F, (I?), J with respect to some of the thermal areas of the S. aquatilis Sauvageau (62) ...... (45-50 C) world is indicated by the letters in capitals. The E, F, G, H, J locations of these are shown in Fig. 3, and relevant (B?), references are given there. Aphanocapsa thermalis (Kutz.) Brugger (14, 145) ...... (55+ C) have the high nutrient characteristics of eutrophic A, B, D, E, F, G, I, J waters, but these are often countered by poi- Order: Chamaesiphonales Family: Pleurocapsaceae sonous concentrations of certain elements, low Pleurocapsa [? minor Hansg. (145)]... (52-54 C) pH, or extremely high salinity. It is probable A, G, (H?), J that many thermal streams cause eutrophication in the bodies of water into which they drain. Order: Oscillatoriales Family: Oscillatoriaceae Harrington and Wright (96) have measured a Oscillatoria terebriformis Ag. "thermal- red" (53) ...... 53 C considerably higher standing crop and primary A, (J?) (B, D, E, F, G, and I-other productivity in the Firehole River below the forms ?) thermal tributaries from the Upper Geyser Basin 0. animalis Ag. (73) ...... (?55 C) (Yellowstone) than above. Similarly, the chemis- A, D, E, F, G, J try or temperature of hot spring effluents appar- 0. amphibia Ag. (62) ...... (57 C) ently causes increases in insect biomass in the B, D, E, F, G, J Gibbon River of Yellowstone (178). If poisonous 0. geminata Menegh. (62, 145) ...... (55 C) concentrations of some elements were present in A, B, D, E, F, G, H, I, J the thermal streams, these would be diluted to o okenii Ag. (116) ...... (60+ C) (B?), E, F, H, J subtoxic levels in major freshwater streams. 0. tenuis Ag. (unpublished data) ...... 45-47 C The dissolved organic compounds present in B, C, D, E, F, G, H, I, J thermal streams have seldom been measured. Spirulina sp. [incl. S. labyrinthiformis Brock (30) found 2.4 mg of organic carbon/liter Gom. (unpublished data)] ...... (55-60 C) both in the source pool and at the foot of the A, B, F, H, J drainway of his main research spring. The con- Phormidium laminosum (Ag.)Gom. centration of organic solutes and their identities (54) ...... 57-(60) C should be of considerable interest, since little [? incl. P. tenue (menegh.) Gom. and P. valderianum (Delp.) Gom.] is known about the natural energy or carbon A, B, C, D, E, F, G, H, I, J sources for bacteria that grow at very high tem- peratures [75 to 95 C (26)]. Brock and Freeze 486 CASTENHOLZ BAcrERIOL. REV. (42) have recently described a new aerobic, centimeter. On the other hand, Setchell (158) nonsporulating bacterium (Thermus aquaticus) and Nash (139) reported blue-green algae in isolated by high temperature (70 to 75 C) enrich- Yellowstone at temperatures no higher than ment from many hot springs. This may represent about 73 and 75 C, respectively, which agrees one of the important heterotrophs of hot-spring essentially with the observations by Brock and mats. myself. Kempner (113) measured 32p incorpora- The undermat of carotenoid-containing fila- tion into nucleic acids in a few adjacent springs mentous bacteria (presumably heterotrophs) in Yellowstone and found none above 73 C; he must also depend on organic solutes either from concluded that this temperature represented the the spring source or from excretions or lysates upper limit of life. Probably he was working with of the photoautotrophs of the upper layers. In waters that contained Synechococcus but not some springs (e.g., Kah-nee-ta, Oregon) thick the bacteria which grow at considerably higher top mats of orange filaments (about 1.5 /Am temperatures ( >90 C) in nearby springs (26, 29). wide) predominate in summer over large areas Some of the discrepancies referred to might at temperatures ranging from about 55 to 40 C. be resolved if the duration of exposure to the The specific energy and carbon sources for growth temperature in question were known. Some of are unknown, but chlorophyll-a is lacking. the reports of blue-green algae above probable maximum constant growth temperatures may DISTRIBUTION OF SPECIES represent short-term exposures to the higher Upper and Lower Temperature Limits temperature. For example, M. laminosus was found in Iceland in a particular spring that varied The maximum constant temperature that will from about 60 to 70 C approximately every 20 sustain growth under natural conditions is still min (54). What is thought to be the same organ- unresolved for most cyanophytes, but 73 to 74 C ism was isolated from the spring, cultured, and may be the upper limit for the highest tempera- cloned. In the laboratory, its maximum constant ture type [Synechococcus (Table 3)]. Brock (31) growth temperature was 63 to 64 C. However, has demonstrated photoincorporation of 14C- it could withstand about 9 hr of exposure to 70 C carbonate in Synechecococcus populations at without noticeable mortality (54). temperatures as high as 73 C in a Yellowstone In North American springs, 74 C is about the hot spring. In complex hot-spring medium, S. maximum constant temperature at which one lividus has a generation time of less than 24 hr at or more species of Synechococcus occurs (Table 70 C (146). There are numerous reports of blue- 3). Where these species do not occur or where green algae occurring at higher temperatures chemical conditions may be inhibitory, the upper (see 27, 52, 59, 145, 179, 187). Copeland (59) temperature boundary for photoautotrophs is listed five species above 75 C in Yellowstone, lower. In hot springs of the Cascade Range in one as high as 85 C. Mann and Schlicting (131), Oregon, 64 to 68 C is approximately the upper more recently, reported similarly high tempera- limit; unicellular Synechococcus is usually the ture ranges in Yellowstone. Brock (29, 32) and only cyanophyte above 58 C. Farther to the east I (52) have examined some of the same springs in Oregon, and in Yellowstone, 73 to 74 C is the and have concluded that the organisms in apparent boundary in alkaline springs. In acid- question are probably flexibacteria or other sulfur springs below pH 4 in Yellowstone, no nonchlorophyllous filamentous procaryotes, not cyanophytes occur; C. caldarium is the only photosynthetic cyanophytes. Although fluores- photoautotroph over 45 C and it extends to about cence microscopy may not be capable of resolving 57 C (16, 34). In Iceland, the photosynthetic very small quantities of pigment, Brock did not upper limit in alkaline springs is about 63 C. find any indication of chlorophyll-a in the mate- This is a reflection of the absence of all species of rial collected (32). Organisms of these higher Synechococcus and the presence of M. laminosus temperatures are sometimes brightly colored, which tolerates this temperature in many alkaline presumably because of carotenoid pigments. So springs throughout the world (Table 3; 54, 154). far, there has been no demonstration of photo- Similar reasons for depressed upper-temperature synthesis or growth of photoautotrophs over borders probably apply to New Zealand (139) 73 C. Most of the reports of higher tolerances and many other thermal areas lacking high- (particularly numerous in the older literature) temperature races of Synechococcus. are based simply on presence; and the tempera- The lower temperature limits for growth are ture, even if measured at the precise site of the very poorly defined. Most of the thermophilic specimen, was probably taken with a bulb ther- cyanophytes which have been examined in this mometer in water, which may often stratify to respect grow poorly or not at all below 30 to 35 C. the extent of a 10-degree difference in a vertical Certain races of Synechococcus will not grow VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 487 statements about this, primarily because the 100 specific identification of blue-green algae is so uncertain at present and few thermal areas have been examined exhaustively. Nevertheless, I have chosen several of the species of blue-green

<60 algae that are most frequently reported from hot springs above 45 C. These are included in Table 3 together with some general information on their thermophile of the hot springs and generally to have very wide geographical distribu- the marine type. However, information of a tions. However, it is difficult to make categorical genetic type, based on deoxyribonucleic acid 488 CASTENHOLZ BAcIERIO. REV. hybridization studies, base composition, and confident that the same species occurs in all these other characteristics may soon revolutionize the localities. classification of blue-green algae (60, 68, 130). At Iceland probably has only about six or seven present, to avoid the confusion of organisms by blue-green algae capable of growing above 45 C, oversimplification, it is important that the more whereas hot waters of Europe, Asia, and North narrowly descriptive names be used, with Drouet's America generally support at least twice this name in addition, or vice versa. number of species (54). Iceland has been active Besides being the most ubiquitous and cosmo- thermally throughout the last glaciation; but politan thermophile, M. laminosus (Ag.) Cohn hot waters were probably covered by the ice until and its several "forms" is one of the easiest about 9,000 years ago. Relative youth and the species of blue-green algae to recognize because rarity of dispersal from other thermal areas are of its unusually high degree of morphological probably major factors in accounting for the complexity (27, 85, 154). Some of the morpho- impoverished condition of the Iceland flora (54). logical forms may simply be developmental The range in chemistry of thermal waters in phases or temperature-controlled manifestations Iceland is probably great enough to support any of the same organism. Preliminary experiments of the earth's thermophilic blue-green species have indicated, however, that there are at least (17, 174, 181; Table 2). The species in Iceland two genetic "races" with unique morphologies all appear to be cosmopolitan. and temperature tolerances (54). Although the Some thermophilic species and varieties are literature has not been exhaustively searched, it broadly endemic, but are absent from a few is evident that M. laminosus has been reported geographical regions (see Fig. 3). For example, in thermal waters (pH 5 to 9+) from almost all the narrowly elongate species of Synechococcus parts of the earth (154), with the possible excep- (e.g., S. elongatus) which inhabit waters to about tion of some midoceanic islands [e.g., the Azores 70 C in North America, Japan, and the eastern (37)] or recently deglaciated areas [e.g., west Mediterranean are absent from parts of western coast of Greenland (27)]. In regions of many hot Europe, Iceland, and possibly New Zealand. S. springs, such as Yellowstone Park, some springs lividus, as such, has only been reported from are dominated by Mastigocladus, whereas others North America and Japan (71, 187). Narrower (superficially similar chemically) are almost com- distribution patterns, such as that of Oscillatoria pletely dominated by Synechococcus. Throughout terebriformis, probably occur, but may represent the world a "Mastigocladus Spring" appears to the distribution of acceptable habitats rather be a recognizable entity. No one, however, has than actual momentary stages of geographical fixed the reasons for this discrimination by spread. 0. terebriformis apparently occurs only habitat. From the little that is known about in restricted hot-spring regions of western North Mastigocladus, other thermophiles, and hot America. The same red-brown thermophilic spring chemistry, it is tempting to suggest that type may also occur in Japan (155), but doubt- spring sources deficient in combined nitrogen fully in other continents, although a number of will allow the N-fixing Mastigocladus to dominate different genetic entities fit the broad description from about 60 to about 50 C, since neither (Agardh ex Gomont) bearing this binomial. Synechococcus nor other thermophiles of this "Thermal-red" 0. terebriformis seems to occur temperature range are apparently capable of in almost all alkaline springs of western and nitrogen fixation (168). On the other hand, eastern Oregon, northern California, and Nevada nitrate- or ammonium-rich springs may allow (53). However, it has not been found by me or other thermophiles to compete successfully Brock in the diverse springs of Yellowstone against Mastigocladus, perhaps through more National Park, which are merely 600 miles from rapid growth rates. There are also reports that abundant populations in Oregon. The 0. terebri- Mastigocladus is less tolerant of hydrogen sulfide formis referred to by Stockner (169) in tepid than other species in hot springs (see reference springs at Mt. Rainier is nonthermophilic, green, 154). Other chemical constituents could also be and with different cell dimensions in culture involved in the exclusion of Mastigocladus from (unpublished data). some springs. Most remote thermal areas have not been Problems of Survival and Transport investigated with regard to the species present. The capacity of thermophilic cyanophytes to Phormidium laminosum is alleged to be a common survive outside the normal habitat has been little companion of the cosmopolitan Mastigocladus. studied. None of the extreme thermophiles pro- However, specific differentiations in the narrow- duce typical resting spores. Certain species, such trichome (<3 Am) members of the genus Phor- as M. laminosus and P. laminosum, are very midium are so uncertain that one cannot be tolerant of freezing and drying, whereas S. lividus VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 489 is less so and 0. terebriformis is very labile. The apparent deterioration at high temperatures. transport of thermophiles over long distances These mats are largely composed of living micro- would probably require drying or freezing, or organisms, held together by abundant extra- both, during the journey if the cells were located cellular gelatinous materials. The cyanophytes on the outside of the carrier. Proctor et al. (149) usually occupy only the first few millimeters or demonstrated that freshwater algae, other cypto- less, and orange, yellow, red, or flesh-colorea gams, and small invertebrates are transported filamentous bacteria make up the rest. Although internally by various water birds, but few blue- unicellular bacteria are also present, and become green algae were found (148). abundant in organic enrichments (31), carote- In hot springs of equatorial or temperate noid-containing flexibacteria may be the principal latitudes, a dormancy period is not required for heterotrophs in many hot springs. The gliding winter survival. In Yellowstone and Oregon hot movements of these bacteria (and flexional springs, midday light intensities in winter often movements of some) together with slight simi- exceed 0.2 cal per cm2 per min [400 to 700 nm larities in carotenoid pigments (1, 84) indicate (88)], and blue-green algal populations similar that they may be nonchlorophyllous homologues to those of summer are maintained in the steep- of blue-green algae. The leakage or lytic products ened thermal gradient (39, 41, 52, 53, 145, 169). of the primary producers are probably the main In Iceland, however, the 2 months of near total energy sources available to the bacteria, although darkness in winter implies a dormant phase for the source waters of some hot springs may contain the cyanophytes, since they are probably obligate some organic solutes derived ultimately from photoautotrophs. The hot-spring mats disinte- surface waters. The bacterial undermat is essen- grate and wash out (54, 153, 175). The stress of tially the closed end of energy flow. This material prolonged darkness may also limit the number does not readily decompose except when washed of species to some extent, although species unable downstream into nonthermal waters. to withstand these conditions probably would It was suggested earlier by Schwabe (157) and not easily survive transport to Iceland initially by Brock (30) that the organic accumulation in (54). 0. terebriformis ("thermal-red") of North thermal waters is due primarily to the general America may be such a labile species. absence of herbivores above 43 to 48 C. This One of the interesting questions is whether, may be difficult to demonstrate directly, but what in the dispersal of thermophilic blue-green algae, would appear to be the effect of animal grazing the carrier must transport the inoculum from is quite apparent below certain temperatures in one hot spring to another or whether sources of numerous hot springs. In many cases (Oregon, inoculum are available elsewhere. Thermophilic Iceland, and Yellowstone) the mat is corrosively spore-forming bacteria (e.g., B. stearothermoph- dissected by burrowing ephydrid (diptera) larvae ilus) can be isolated from common soils as which may reach large numbers below above well as from hot springs (31). Viable cells of 45 C, although specific predators such as doli- obligate blue-green algal thermophiles (at least chopodid flies are also present at some springs extreme types) may not be found far from thermal (28, 175). In some Oregon hot springs, a thermo- areas, but no one has really looked. I have occa- philic ostracod forms very dense populations sionally grown blue-green algae out of samples up to about 47 to 48 C without any apparent of common soil in the Willamette Valley (Ore- competitors or predators present. In such situa- gon), with incubation at 45 C, but I have never tions, there is an abrupt change both in the algal had positive results at 55 C. species and total mass below the maximal tem- perature of these animals. There is generally no STUDIES OF NATURAL POPULATIONS mat accumulation at all. Instead, there may be Mat Formation and leathery nodules composed of the cyanophytes Stability Pleurocapsa sp. and Calothrix sp. These nodules In the thermal range (above 45 C) of nonacid are able to withstand the apparent grazing pres- hot springs, there occur gelatinous or calcareous sure in flowing waters from about 47 to 35 C. mats of various colors, sometimes several centi- Brock et al. (28) have demonstrated by direct meters in thickness (Fig. 1, 2). The topmost layer means, using "IC-labeled blue-green algae and is generally composed of photosynthetic blue- bacteria, that ephydrid larvae and adults feed green algae. A variety of mat structures has been on the microorganisms of hot springs. I have described by Schwabe (155, 156), Brock (33), found that the thermophilic ostracods of Hunter's Stockner (170), and a few others. One of the Hot Springs (Oregon) can rapidly decimate a striking aspects of these mats is the variety of population of 0. terebriformis. In these springs, bright green, brown, red, orange, and yellow the lower temperature edge of 0. terebriformis colors. These large mats accumulate without mats (46 to 48 C) is probably determined by the 490 CASTENHOLZ BAcrERioL. REV. feeding range of this herbivore rather than by a Oregon) where a gel-like translucent orange layer greatly decreased growth rate of the algae at that a few millimeters thick (composed of filamentous temperature (see Fig. 4). "bacteria") lies over a dark green layer of Syne- The grazers of the mats at lower temperatures chococcus. probably aid in decomposition by ingesting Under nearly all types of photosynthetic covers microorganisms and depositing large amounts in nonacid springs lies the orange-to-flesh colored, of partly changed fecal matter which may then heterotrophic undermat. In an early statement, be further decomposed by the microorganisms of Brock (30) suggested that the many layers of lower temperatures. In areas worked over by typical hot-spring undermats represented the dipteran larvae, ostracods, amphipods, and other accumulation of several years. Later, he realized animals, large amounts of characteristically that accumulation is considerably more rapid flocculent fecal matter with associated micro- and that the distinct layering cannot be explained organisms may accumulate in pockets of quiet easily (39). Stockner (170), studying mainly tepid water. This is probably a reliable indicator of calcareous springs at Mt. Rainier, came to a heavy grazing activity. similar realization. In most alkaline hot springs, the blue-green Brock has examined the manner in which mats algal cells are concentrated in the uppermost few are formed (33). He replaced a portion of the millimeters of the mat, although filamentous hot-spring drainway (about 55 C) with a wooden bacteria may also be abundant there. Brock channel 6 inches wide and 7 ft long. An orange (33, 41) has demonstrated by 14CO2 incorporation cover of undermat formed first and was about 2 that photosynthesis is limited to the upper portion mm thick in 3 weeks. This was followed by a top of the mat. For example, in a mat 1 to 2 cm cover of green Synechococcus which first appeared thick essentially all activity was restricted to the as small patches, eventually making a complete top 2 to 5 mm. Chlorophyll measurements made cover in an additional 4 weeks. In hot springs of by spectrophotometry or fluorescence microscopy eastern Oregon, I found that the blue-green alga corresponded with the incorporation results. In was the first apparent colonizer on natural and the top 5 mm of a typical core from 57 C in a plexiglass substrates. It is possible that in some Yellowstone spring, unicellular Synechococcus stretches of thermal streams organic solutes are was essentially the only photoautotroph (41). at such low concentrations that bacterial growth This upper portion was subdivided into four depends on intimate contact with photoauto- separate layers for autoradiographic analyses trophs. During optimal periods, a new covering after photosynthetic incubation. In the uppermost of Synechococcus will coat plates (15 by 15 cm) 1.2 to 1.3 mm, 95% of the cells were labeled, in a few days, and a typical orange undermat will whereas only 35% (of fewer cells) in the lowest be visible a few days later. The orange undermat sublayer were labeled. Furthermore, the top increases in thickness by growth and not by cells averaged 4.8 silver g per cell and the lowest immigration. only 1.3 per cell, indicating the obvious light In eastern Oregon hot springs, new mat mate- limitations at depth. 0. terebriformis ("thermal rial on denuded substrates and submerged plexi- red") is ubiquitous in Oregon springs below glass plates increased in thickness at a rate of about 54 C (Fig. 1, 2). It forms deep red-brown 1 to 2 mm per week. Commonly, this corre- mats 2 mm in thickness and so dense that com- sponded to values of about 1 to 2 g of organic monly only about 1% of the incident light is matter (ash-free dry weight) per m2 per day and transmitted (53). Since these mats are motile and 5 to 10 mg of chlorophyll-a per m2 per day. tend to override immotile Synechococcus surface These are not true growth rates since they exclude covers, the latter become severely shaded. Even- losses from decomposition, grazing, and washout tually the Synechococcus may disappear (53). In and include the possible addition of cells recruited North America, the photosynthetic layer above from upstream. Cells of S. lividus are almost about 60 C is dominated by narrow-cell forms bacterial in size (5 by 1.5 Mum) and settle out very of Synechococcus resembling S. lividus Copeland. slowly even in unshaken flasks. The ability of a In Yellowstone, Oregon, and other hot springs dislodged cell to successfully recolonize will surely there are great variations in photosynthetic diminish with increased flow rate or steepness of covers below this temperature. The photosyn- the thermal gradient. In a few cases, however, thetic top layer may consist of densely bound or whole masses of detached mat from upstream unbound populations of unicells (e.g., S. lividus, became lodged on new substrate and obviously S. minervae), compact masses of gel-bound or are added to the yield. The cause of the very free filaments (Phormidium sp., Oscillatoria sp.), distinct layering of most mats is not understood or tufts of branching filaments (e.g., M. lamino- (Fig. 2). It occurs in mats of flowing streams and sus). There are also situations (e.g., Kah-nee-ta, nonflowing thermal pools. If the specific organ- VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 491 isms of the various layers of undermat were noted (53). In addition, the upper edge of the known, the reasons for layering might be more Oscillatoria mat will advance upstream (by mass obvious. The steepness of the vertical oxygen gliding of trichomes), keeping pace with the 54 C gradient in the mat should also have an important point when the temperature drops because of influence on the type and growth rate of these. wind or decreasing air temperature. Most of the In the Oregon springs, there is often a vivid movements of Oscillatoria described are probably pink-red layer of Chromatium (thiorhodaceae) adaptive in that they enable the cells to maintain immediately under 0. terebriformis at tempera- themselves close to optimal light and temperature tures below about 57 to 60 C. If the Chromatium conditions which change diurnally and seasonally. were active photosynthetically, the local condi- The mat of moving trichomes also allows little tions should be nearly anaerobic and at greater competition within its optimal temperature depths in the orange undermat also, at least range, since it is always able to form the topmost above the temperature of extensive burrowing layer. The common, more sedentary cyanophyte by dipteran larvae (about 45 C). Species of covers of hot springs have no adequate way to Saprospira isolated from hot springs by Lewin cope with diurnal fluctuations in temperature (126) showed aerobic tendencies, but it is not and light except by physiological tolerance. In known whether these were important components many cases, these types are not able to colonize of undermats. Similarly, the nonsporulating or grow rapidly enough in an upstream direction bacterium T. aquaticus, isolated from Yellowstone to keep pace with the steepening of the tempera- springs by incubation at 70 to 75 C, is an obligate ture gradient with on-coming cold weather. As a aerobe (44). consequence, many thermal mats are left below the growth temperature range in winter when they Movements of Filaments and Mats bleach, disintegrate, and wash downstream (53). In most hot springs in Oregon [and apparently Over wintering is probably by vegetative or in Japan also (155)], top layers of highly motile metabolically inactive cells along the upstream 0. terebriformis are common (Fig. 1, 2, 53). 0. edges of the springs. terebriformis forms dense mats, commonly over 2 mm in thickness, which consist of free-moving Measurements of Photosynthesis and but intricately interwoven trichomes with little Growth or no gel binding them. The integrity of the mat In 1966, Thomas D. and M. Louise Brock is maintained by the intricacy of the trichome determined the standing crop of algae and other entanglement, and the position of the whole mat material along a thermal gradient in alkaline hot in flowing water is probably maintained by the springs in Yellowstone and Iceland (35) by meas- penetration of free trichome ends into the sub- urements of chlorophyll, ribonucleic acid, and strate. Almost all other cyanophyte covers in hot protein (36). In the Yellowstone drainway all springs are held together and kept in place by three materials were produced at maximal levels the gelatinous material formed by the blue-green at a temperature between 51 and 56 C. This was algae, the filamentous bacteria, or both. 0. also the region in the gradient with the greatest terebriformis mats, where found, dominate the mat accumulation as determined visually. Com- thermal gradient from about 54 C down to about mon values for chlorophyll were 700 to 800 mg/ 48 to 35 C, the lower edge apparently depending m2 (35). In the single Icelandic spring examined, on the abundance of the herbivorous ostracod all three materials were maximal at a temperature described earlier. of about 48 C. It has already been mentioned During midday periods of high light intensity, that the species may be different in the two 0. terebriformis mats may contract to only a regions, and that the upper temperature limit of small fraction of their area under low light or photoautotrophs in Iceland is only about 63 C. darkness, exposing covers of Synechococcus or In both cases, however, the standing-crop maxima the orange undermat of filamentous bacteria. were above the temperature range of herbivores Mat-edge retraction rates may be as rapid as 100 or any other animals in the springs. Positions of cm/hr. Trichomes may migrate downward into maximal mat accumulation do not necessarily some substrates en masse (53). With light reduc- fix the locations of the maximal photosynthetic tion they will rapidly return to the surface and or growth rates of blue-green algae or the hetero- spread out by gliding. Mass aggregations and trophs. Since the first paper, the Brocks have contractions were described in Oscillatoria (51, published several additional reports, most of 53), and a somewhat similar phenomenon in a which are concerned with natural rates of photo- nonthermophilic Anabaena (180). Contraction synthesis in Yellowstone hot springs (30-33, 38- of the mat edge in 0. terebriformis in response 41). Their principal experimental stream was to supraoptimal temperatures has also been Mushroom Spring in the White Creek drainage 492 CASTENHOLZ BACTERIOL. REV. of the Lower Geyser Basin, a few hundred meters the external light intensity. In a later study, the from Great Fountain Geyser. The temperature Brocks found that there was seldom any indica- for the maximal crop of chlorophyll was again tion of photosynthetic inhibition of algal mats between 50 and 60 C (30). The incorporation of even by the highest light intensities [about 0.56 '4CO2 per unit area was also highest in the general cal per cm2 per min, corrected for 400 to 700 nm area of maximal algal standing crop, as expected. by a factor of 0.4 (see reference 172)] unless the Incorporation studies were performed with mat populations had been adapted to reduced light cores set upright in horizontally positioned vials. intensities previously (41). Photosynthetic light These cores were generally incubated with saturation, however, occurred in some cases at NaH'4C03 in spring water at original tempera- values of about 0.3 cal per cm2 per min (my tures for 1 hr. Although objections might be estimate for station I, 71 C, corrected for 400 to raised about possible stress conditions in the vial 700 nm) which was about 50% of full midday (170), the time is short, and there has been no summer insolation. Brock (in Yellowstone) and indication yet that thermophilic blue-green algae I (in Oregon) have noted that mats of Synecho- have a requirement of high-water velocity which coccus at the higher temperatures (i.e., 68 to 74 C) has been attributed to some algae of streams appear thinner than at lower temperatures. Thus, (184). it is probable that a greater percentage of the Brock (30) also plotted the photosynthetic cells in these thinner mats become light saturated, rates on a per chlorophyll basis to determine resulting in a relatively high photosynthetic photosynthetic efficiency. The greatest efficiency efficiency expression. An increase in apparent occurred at temperatures between 40 and 50 C efficiency near the upper temperature limit was instead of in the region ofthe greatest chlorophyll. indeed found by Brock in one study (30). How- It is often useful to express photosynthesis on ever, the Synechococcus mats at these high tem- some basis other than area, but in this case, the peratures may be thinner because of poorer use of pigment as a basis must be distinguished photosynthetic and growth rates for individual from the same expression in laboratory studies light-saturated cells than at somewhat lower (see reference 167). Since intact cores of the algal temperatures. If this is true, the higher efficiency mat were used in the experiments, all photo- expression would be rather misleading in terms synthetic cells were compacted in the upper few of physiology. millimeters. Nevertheless, even in this short Additional information has come from two vertical distance there is probably a steep drop experiments in Oregon hot springs (unpublished in light intensity along with an increase in individ- data). In these springs, Synechococcus cells can ual cell chlorophyll (see references 44, 137). The be easily removed by suction from the mat surface photosynthetic rate at the top may have been since they are not bound in place by mucilage as light saturated (or even inhibited) at the same in many Yellowstone springs. Thus, dilute cell time that cells at greater depth were severely suspensions with a minimal degree of self-shading limited by dim light. Thus, the expression of can be prepared for photosynthetic studies, photosynthetic rate per unit of chlorophyll in although the cells used may still have quite mixed the case of the field experiments differs in meaning light histories. Using methods of 'IC incorpora- from the same expression in laboratory studies tion and analysis similar to those of Brock, I where it would be assumed that all of the cells measured the photosynthetic rates of replicate are in suspension, equally exposed to light, and suspensions of Synechococcus for 30 min in two contain the same amounts of chlorophyll. In Oregon springs under different light intensities, the field, it should be expected that photosyn- simultaneously. In both experiments, the cells thetic efficiency (of intact mats) will be less in were collected from a temperature of 60 C and regions of thickest algal mat (highest chlorophyll incubated at the same temperature. The results values) because of the greater degree of light showed that photosynthetic saturation for these extinction. cells occurred at about 0.07 to 0.08 cal per cm2 The photosynthetic characteristics of natural per min (400 to 700 nm), approximately 1,500 to algal mats have been illustrated in several studies 2,000 ft-c. Intensities of about 0.2 cal per cm2 by the Brocks. In one experiment, Brock showed per min were still in the saturation range. How- that the photosynthetic rate per unit of chloro- ever, an intensity of about 0.4 depressed photo- phyll was dependent on light intensity throughout synthesis to about 50% of the maximal rate. the course of a single day in summer (30). This Although the conditions and the species strains can be best explained by assuming that light used might be different in Oregon and Yellow- attenuates greatly within the algal layer, leaving stone, comparisons are of interest. Brock's finding many cells photosynthetically unsaturated; the (41) of photosynthetic saturation at about 0.3 percentage of cells saturated would depend on cal per cm2 per min in the thinner mat at 71 C VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 493 would indicate that mat thickness is still a factor self-shading, since the number of productive hours at intensities lower than this, if we accept the changes with day length, even though photo- evidence that indicates individual cells are satu- synthesis may be saturated during many of the rated at intensities as low as 0.07 cal per cm2 per daylight hours. min. Thicker mats at lower temperatures should probably not show photosynthetic saturation CULTIVATION OF THERMOPHILIC even under the brightest natural light. CYANOPHYTES The Brocks (38) measured steady-state growth rates of natural populations of Synechococcus by There is little information on the cultivation novel methods. On a hard-bottom drainway of of thermophilic blue-green algae. The following sinter in thermal the section will summarize primarily the methods siliceous shallow streams, which have been used in my laboratory. Most complete darkening of a Synechococcus mat media were originally designed for nonthermo- apparently stops growth completely, as would philic cyanophytes (7, 108) and later modified be expected in the case of obligate photoauto- and used with thermophiles (67, 69, 105). Media trophs. In the dark, washout of cells appears to and methods for unicellular types, including the have continued at a normal rate until eventually semithermophilic Anacystis nidulans, have been the sinter was bare. The loss rate was exponential. developed by Van Baalen (177) and Allen (10). A The time when the population was 50% of the more original size (as determined by cell count) was detailed description of the culture tech- equated with the generation time required to niques and of the culture collection of thermo- maintain the observed steady-state population philic blue-green algae at the University of under photic conditions. The estimates of genera- Oregon will be published in Schweizerische tion time were 40 hr at the 70 C station and 22 hr Zeitschrift fur Hydrologie during 1970 (55). at the 72 C station. The rate of increase in cell numbers can generally be equated safely with the Medium and Nutrition rate of increase in mass over a 24-hr period, in the Many of the predominant photosynthetic blue- case of unicellular organisms. Assuming that green algae of hot springs grow vigorously in mass increase is possible only during the 15-hr mineral medium. Species from hot springs of light period, I have shortened the Brock values many types and a variety of geographical regions for generation times to 25 and 13.7 hr, respec- are grown on defined medium D, with minor tively. This means that during the light period the modifications (Table 4). Complex media with rate would be 1.0 and 1.7 mass doublings per day. appropriate hot-spring waters have also been These rates are very close to the maximal rate used extensively for some species (Table 4; 146). (about 2.0 mass doublings per day) obtained for a In all cases, medium D has worked as well or high temperature strain of Synechococcus in my better. The only organic compound in the laboratory at 65 C, under continuous light and a medium is the metal chelator, nitrilotriacetic variety of intensities, gas mixtures, and agitation acid (NTA). Although NTA may be used as a rates (J. Meeks, unpublished data). Maximal C and N source by Escherichia coli (83), it is rates at 70 C are consistently somewhat lower probably not metabolized by cyanophytes. The and even more comparable with the Brock field medium contains all of the nutrients known as values. The field method described should be very requirements for blue-green algae (92, 104, 142). useful in springs or streams where flow rates are Nitrate is probably the sole source of combined constant and where unicellular algae predominate. nitrogen in the unmodified medium, although Difficulties would arise if components of the algal NTA may conceivably provide nitrogen to some mat were facultative dark heterotrophs, if dark- organisms. M. laminosus and Calothrix sp. are ness affected the tenacity of cell attachment to the the only thermophilic blue-green algae for which substrate, or if washout was erratic and involved nitrogen fixation has been reported (78, 152, 168). chunks or tufts of mat rather than individual cells. Auxotrophy is known in relatively few cyano- Stockner (170) has estimated primary produc- phytes (104) and not at all in thermophilic species. tion by measuring diurnal changes in dissolved Although photoassimilation of organic com- oxygen and also increases in organic matter on pounds, such as acetate, may be common in denuded substrates in the drainways of tepid cyanophytes (101, 143), only a few blue-green (37 C) springs at Mt. Rainier, Wash. He found a algae will grow heterotrophically in the dark (4, correlation between blue-green algal production 75, 104, 114, 122, 162). Among photosynthetic and monthly light intensity-light duration thermophiles, only the eucaryotic alga C. calda- means (gram calories per square centimeter per rium is known to be a heterotroph (6, 16). day) throughout most of the year, even summer. The salinity of the great majority of neutral- This may be expected, however, with or without alkaline hot springs is much higher than that of 494 CASTENHOLZ BACTERIOL. RiEV.

TABLE 4. Composition of media and solutions in a variety of complex and defined organic Medium Da culture media (5), as well as in relatively simple Double distilled water ...... 1,000 ml medium composed of distilled water containing Nitrilotriacetic acid...... 0.1 g 0.08% MgSO4.7 H20, 0.018% CaCl2-2 H20, Miconutrient solution...... 0.5 ml 0.05% KNO3, 0.28% NH4Cl, 4% glycerol, 0.6% FeC13 solutions...... 1.0 ml sodium glycerophosphate, and 1% glucose mono- CaSO4-2 H20...... 0.06 g hydrate, which was adjusted to pH 6.8 and MgSO4.7 H20...... g0.10 sterilized by filtration through a membrane filter NaCl...... 0.008 g (47). KNO3 ...... 0.103 g NaNO3...... 0.689 g Isolation and Maintenance ...... g Na2HPO4 0.111 Algal material collected from hot springs should Micronutrient Solution be quite dilute in sample vials containing spring Distilled water ...... 1,000 ml water. Most keep very well, often for several H2SO4 (concd)...... 0.5 ml weeks or months, in the dark at a temperature MnSO4 H20...... 2.28 g between 10 and 20 C. A temperature below 5 C is ZnSO4 7H20 ...... 0.50 g quite detrimental to some. However, even 20 to H3B0O...... 0.50 g 30 C in darkness will generally not harm samples CuSOc45H20 ...... 0.025 g for several days if kept dilute. Therefore, no Na2MoO4-2H20...... 0.025 g CoCl2 6H20 .0...... O 045 g special precautions are required for shipment except padding and darkness. Complex hot spring water medium Normal D medium (Table 4) with a pH of 7.5 Hot spring water ...... 1,000 ml after autoclaving is used as almost a universal Soil extractc...... 50 ml medium for the isolation and growth of cyano- Ethylenediaminetetraacetate-Fe phytes from hot springs which range in pH from (13% Fe) ...... 0.005 g 5 to over 9. Most thermophilic blue-green algae, MgSO4- 7H20 ...... 0.100 g as well as the great majority of mesophilic types, KNO3 ...... 0.260 g K2HPPO4...... 0.100 g prefer an alkaline environment (104). The pH of D medium rises as a result of algal growth, as it a Designed by R. P. Sheridan, medium D is is essentially unbuffered. Medium D of pH 4 has prepared as a 10-fold concentrated stock and been used to isolate and cultivate the thermophilic stored at 4 C unautoclaved. The pH is adjusted eucaryote Cyanidium caldarium from acid-sulfate after dilution (double-distilled water) to 8.2 with hot springs. 1 M NaOH. The final pH is 7.5 to 7.6 after the auto- Collected material may be handled in several claved medium has cooled and cleared completely. b FeCla solution consists of 1,000 ml of distilled ways in the laboratory. Filamentous types which water and 0.2905 g of FeCl3. are motile or have motile hormogonia (e.g., c Approximately 400 g of brown loam is auto- Oscillatoria, Phormidium, Lyngbya, Symploca, claved in 1,000 ml of water for 40 min. Mixed Calothrix, Mastigocladus) will generally isolate slurry is filtered through triple-layered Whatman themselves by gliding away from the inoculum no. 1 paper while warm. Clear amber liquid results. particle in a few to several hours on D medium or Cg-10 medium (177) solidified with 1.0 to 1.5% most surface waters. Nevertheless, some essential agar in plates and incubated at an appropriate nutrients such as combined nitrogen or phosphate temperature. Blocks of agar with single trichomes may be very low. However, none of the thermo- may then be picked off for inoculation into liquid philes in culture seems to be sensitive to high or agar medium, in one step establishing clones levels of these elements, unlike obligate oligo- which may also be axenic. Some thicker tri- trophs (see reference 82). Nevertheless, there are chomes (e.g., 0. princeps) penetrate into and some very common thermophilic blue-greens that glide well through agar medium, and there is an have not responded to my isolation techniques. even greater likelihood in such cases of freeing the These include the varieties of S. minervae and trichomes of bacteria. With some slow-moving species of Spirulina. species the original inoculation on agar medium Medium D, with the addition of 0.1% tryptone may be incubated for a few to several days. This and 0.1 % yeast extract, was also used successfully will eventually establish a peripheral mass of (46 to 79 C) for the culture of the nonsporulating trichomes for subsequent transfer to new agar. bacterium T. aquaticus, obtained from hot springs A gross inoculation on agar medium may also and hot tap water (42). Sporulating, thermophilic be used for nonmotile cyanophytes, but not if bacteria (e.g., B. stearothermophilus) from hot there is an abundance of motile types in the springs and many other sources have been grown sample. The almost ubiquitous M. laminosus and VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 495 P. laminosum have often been separated by dilu- temperature Synechococcus, however, since cells tion streaking. Unicellular forms, such as Syne- may not generally survive either the pick-up by chococcus sp., may also be isolated in this manner, pipette or the transfer to fresh or conditioned especially on the Cg-10 agar of Van Baalen (177), medium. The picking up of agar blocks with and probably on the unicellular medium of Allen single cells that have been sprayed onto the as well (10). Mastigocladus, however, is some- surface of a plate may prove less damaging. We what unique among thermophilic types in its have had some success by inoculating liquid D ability to come up in agar stabs or in agar shake medium with a small amount of a very dilute cell cultures. In the latter method the inoculum is suspension. In new medium, the cells will settle washed, blended, and diluted before dispersing in out and small clonal colonies will appear on the the unsolidified medium at approximately 45 C. flat bottom, but the flask must not be agitated at Axenic cultures of Mastigocladus have also been all during the incubation. Another method has obtained in this way. For the most common been to pull a very dilute cell suspension on to a unicellular thermophiles of North American Whatman GF/C glass-fiber filter, then to place springs, Synechococcus sp., isolations are very the whole disc in liquid D medium. Since these simple for the narrow-form rod types (resembling filters become quite fuzzy when submerged, small S. lividus) which are described as several species colonies may be removed by picking off a sub- by Copeland (59). In springs in which Synecho- merged tuft with a forceps. coccus is present, the bulk inoculation of collected Many larger cells or filaments (e.g., Mastigo- material into liquid D medium usually results in cladus) may be cloned and also purified by a the growth of several blue-green algae. However, series of dilution washes when manually carried with time, narrow-form types of Synechococcus through a series of depression wells, each with almost always outgrow the others and often form sterile medium. A pulled glass capillary tube, a dense suspension bloom. One or more transfers point, or hook may be used under a dissecting of suspended cells generally results in an unialgal microscope for the transfers. Agar plates may also culture. Clones of these rods (1.2 to 2.2 ,um wide) be conveniently used as a working surface for may be established in different ways. For incuba- dragging out and isolating single cells or fila- tion temperatures of 50 C or less, Synechococcus ments from mixed samples (see reference 125). suspensions may be diluted by streaking on the Axenic cultures of some cyanophytes may be Cg-10 agar medium developed by Van Baalen established by treatment with ultraviolet radiation (177) for single cells of A. nidulans. In my labora- (104). With some of thermophilic types, however, tory this medium is modified only slightly: the it appears that the several orange-colored bacteria glycylglycine buffer is reduced to 0.5 g/liter, 1% often present are less sensitive to ultraviolet agar is used, and the pH is adjusted to 8.2 with 1 treatment than the cyanophyte. M NaOH before autoclaving. Colonies from Stock cultures here are usually maintained in single cells of Synechococcus develop on this 125-ml Erlenmeyer flasks with 80 ml of liquid D medium but generally not on D agar. Axenic medium or in test tubes with 15 to 20 ml of cultures of Synechococcus may be obtained in this medium. Incubation temperatures vary with the way. Difficulties arise, however, with the high species. However, the bulk of the cultures are temperature "races" of Synechococcus which kept at 45 C in constant-temperature water baths. will not grow below 50 to 55 C (146). These may The flasks are usually stoppered with nonab- be easily enriched for by incubating the sample sorbent cotton, but at higher temperatures ( >55 in liquid D medium at higher temperatures (e.g., C) Morton metal closures with "fingers" are 60 to 70 C). Material collected above 65 C in most used with Bellco (Bellco Glass, Inc., Vineland, hot springs will contain narrow-form Synecho- N.J.) flasks to slow evaporation. Growth is coccus as the only viable photosynthetic organism. considerably slower in the vessels with closures; However, syneresis of agar or silica gel occurs in some cases this is preferred. The light intensity and the surface dries rapidly at temperatures at the flask level is kept at about 200 to 500 ft-c above 55 C; streak and plating isolation methods (continuous light from Coolwhite fluorescent were unsuccessful. A tedious, but sometimes lamps). The intensity at the culture level in the effective, method of cloning and purification by tubes is often only 50 to 100 ft-c. Both flask and manual isolation has been used (124). It involves tube cultures, particularly when aged, should the spotting of minute drops of a dilute cell never be exposed to high intensities (>600 ft-c) suspension in a film of mineral oil spread on a for more than a couple of hours; death of all cells glass plate or slide. After microscopic examina- may occur. The flask cultures require transferring tions, drops with single cells are sucked up with a about every 4 to 5 weeks, whereas the slower capillary pipette and inoculated into liquid growing tube cultures are generally transferred medium. This has not worked well with high- every 8 to 9 weeks. Agar surfaces are used in flasks .496 CASTENHOLZ BAcTERioL. REV. for maintaining axenic cultures, but agar slants in (117). Dyer and Gafford (66, 67) reported nine tubes dry too quickly to be used at these tem- doublings per day for S. lividus at 52 C, which is peratures. Cotton plugs can best be protected almost comparable to A. nidulans, Chlorella from dust contamination by capping these with a pyrenoidosa TX 7-11-05 at 39 C (166), and *soft porous tissue (e.g., Kimwipes) held with a Chlamydomonas mundana at 33 C (129). Similarly, rubber band. Nonporous caps cause wetting ofthe several isolates of S. lividus from eastern Oregon cotton plugs at these temperatures. Air-circulated springs grew at rates ranging from 8 to 10 dou- incubators at 45 C and above result in excessive blings per day from 45 to 55 C (146). Both A. evaporation; consequently, water baths are pre- nidulans and S. lividus are small unicellular rods ferred. ranging in cell width from about 1.2 to 2.2 Mm. Many thermophilic cyanophytes will tolerate Chlorella and Chlamydomonas are also small, but freezing or drying for a long time. Slow freezing most species or strains do not exceed 3 to 4 (5 to 10 min) to -20 C works with several com- doublings per day (106). Over a fairly wide range mon thermophiles. Holm-Hansen (102) found of phylogenetic groups there appears to be an this method somewhat preferable to rapid freez- inverse correlation between cell size and growth ing, although the blue-greens used were non- rate (see reference 185), but no constant rela- thermophilic types from either polar or temperate tionship between high growth rates and high regions. All culture strains of the ubiquitous M. temperature, although rates of over 3.0 doublings iaminosus and P. laminosum tolerated slow freez- per day appear to be confined to temperatures ing and storage at -20 C in my laboratory with- over 20 C (106). Brock (29) cites somewhat out any apparent cell death, although only shorter generation times for higher temperature Mastigocladus was stored as long as 3 months. species of bacteria than for lower temperature During some trials, various strains of S. fividus types. In thermophilic Synechococcus, the highest tolerated freezing, and some cells were viable temperature races have reduced growth rates after 1 to 2 years of storage, but this method has (146). Whether the observed limit is genetically not worked in every case. 0. terebriformis, a controlled or whether other problems, such as common thermophile in Oregon, will not tolerate nutrient or CO2 availability, are more manifest slow or rapid freezing. Freeze-drying may be at higher temperatures in still unresolved. Current effective with the more labile types, but it has not studies at 60, 65, and 70 C indicate that the yet been tried with these thermophiles (see maximal growth rate under a variety of aerated reference 103). conditions is only about 2.0 doublings per day Dimethyl sulfoxide, glycerol, or other cryo- (J. Meeks, unpublished data). Estimates of protective agents might also enhance survival of growth rates of Synechococcus near 70 C in cells at low temperatures, but they have not been natural populations gave comparable values tested with thermophilic blue-green algae. (38; see earlier section on field results). Dilute cultures and samples of several species Holton (105) estimated a growth rate of about in liquid medium or spring water have been stored 1.5 doublings per day for M. laminosus and con- in complete darkness at about 15 C and have sidered it quite low. It is lower than many blue- survived well for over 4 months. Such a simple green algae but is nevertheless comparable to a method deserves more intensive trials. large number of eucaryotes (106). Nonthermo- M. laminosus will also tolerate dryness at about philic filamentous cyanophytes such as Schizo- 25 C (with or without a desiccant) for at least a thrix calcicola, Anabaena variabilis, Nostoc few months with little loss of viability. Narrow- muscorum, and Tolypothrix tenuis reached form Synechococcus sp. survived, at most, a few maximal growth rates (3.0 to 4.0 doublings per days of drying, and 0. terebriformis does not day) in culture above 30 C (106). The filamentous survive at all. but undifferentiated 0. terebriformis (trichome width, 4 to 6 Mum) approached 5 doublings per Rates of Growth, Photosynthesis, and Respir- day under optimal conditions from 45 to 53 C ation in Culture (Fig. 4). 0. princeps, with the great trichome The remark by Marry (133) that thermal algae width of 20 to 30 + ;Am, has not surpassed 0.5 generally grow slowly was based on the few doublings per day in culture at its 40 C optimum studies available at that time. Since 1961, how- (L. Halfen, unpublished data). Even though in- ever, there have been enough additional reports of formation on growth rates of the blue-green algae thermal growth rates to indicate that such a is still scanty, it seems that the variation among generalization is incorrect. A short generation species is as great in thermophiles as in mesophiles time (about 2 hr) was known in the case of A. and that no clear trends are evident. nidulans, which has a maximal growth rate at 41 C Photosynthetic rates in thermophiles, as but tolerated a temperature no higher than 45 C measured by H'4CO3 or 14CO3 incorporation, VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 497 have been measured mainly in field populations. lection only; rates decreased with temperature However, neither these nor the few laboratory below and above 58.5 C (Fig. 4). In the few measurements have distinguished thermophilic thermophilic cyanophytes that have been studied cyanophytes from the mesophiles or from the bulk in the laboratory, maximal growth or photo- ofthe eucaryotic algae (49, 67, 118, 137, 147, 161). synthetic rates occurred over a considerably Cyanophytes respire aerobically in the dark, greater temperature range (Fig. 4). For this but sometimes at rates lower than those of reason, I suspect that the sharpness of Brock's eucaryotic algae (24, 43, 104, 105, 118, 182). optimum temperature peaks is not real and that In the thermophilic M. laminosus, Holton (105) the species involved have the potential to photo- found a dark endogenous respiratory rate [juliters synthesize and grow at maximal rates over a of 02 uptake per mg (dry weight) per hr] of 8.0 considerably wider range, given time for ac- for actively growing material at 45 C, as com- climation. Little is known about the resilience of pared to 7.5 for active cells of A. nidulans at 39 C thermophilic algae subjected to temperature (118). Biggins (24) found somewhat lower rates changes. Several strains or races of S. lividus (4 to 6 Mliters of 02 per mg per hr) for A. nidukdns were able to tolerate abrupt changes within their at 25 C (a suboptimal temperature). The existence growth range or below it without apparent of peculiarities in respiration, oxidative phos- damage (145, 146). Lag periods in the growth of phorylation, and other aspects of "dark metab- some clones usually varied from a few to several olism" in blue-green algae has been suggested and hours, the time being somewhat proportional to argued recently by several persons (19, 24, 104, the temperature difference involved. A displace- 107, 123, 144, 165). ment from a suboptimal temperature (e.g., 30 to 40 C in clone 53) to a higher temperature in the RESPONSES TO TEMPERATURE AND optimal range (45 to 55 C) generally resulted in a LIGHT INTENSITY shorter lag ( <10 hr) than when the direction was reversed (15 to 72 hr). The ability to withstand Optimal Temperature and Light Intensity considerable changes in temperature is also known The optimal conditions for growth may seldom in obligate thermophilic bacteria, but rapid dis- be described in terms of a single factor. Neverthe- placements to higher temperatures are often fatal less, this has been attempted, using temperature to facultative thermophiles (see reference 21). alone, nutrient concentration alone, light alone, To have a wide range in growth temperature etc. Most of these factors influence the effect of would have considerable adaptive value in a one or more of the others. Thus, experimental thermal stream. procedure can become very complex when one The loss of the ability to grow over the original attempts to consider several factors simul- full temperature range occurs commonly in some taneously. Few workers have done so. The tem- bacteria (74, 141). This is often a loss of low- perature optimum for growth or photosynthesis temperature ability rather than a change in upper of microalgae may be dramatically or subtly limit. Allen (6), using C. caldarium, and Lowen- influenced by light intensity, day length, nutrient stein (128), using Mastigocladus, reported that concentration and availability, CO2 concentra- the maintenance of cultures for long periods tion, pH, constancy of the several factors, and below the optimal growth temperature resulted in preconditioning. In a series of field experiments, a lowering of the upper tolerance limit and of the Brock (31) found that the maximal rate of 14C optimum. Allen had mistakenly assumed that the photoincorporation in natural populations of original temperature maximum for Cyanidium Synechococcus (mainly) taken from several was well above 55 C. Subsequent studies have different temperature regimes was very close to shown that 55 C is very close to the upper limit the temperature from which they had been for this organism in the laboratory (16) or field collected (Fig. 4). For example, sample cores (34). In the case of Mastigocladus, the "main- were collected at 58.5 C in the thermal stream. tenance" consisted of cold storage at a tempera- These were equilibrated for about 5 min in as ture of 5 to 8 C, well below the growth minimum many as 10 new temperatures (about 23 to 73 C) of 25 C (128). Holton (105), while "cleaning up" in glass vials, followed by a 1-hr incubation with his crude material from Laird Hot Springs, B.C., the isotope added. The photosynthetic population which contained M. laminosus, found there was a at 58.5 C in this Yellowstone spring consisted loss in tolerance to high temperature. In isolations almost entirely of Synechococcus sp. Thus, from Iceland hot springs in 1968, I found that by although genetic adaptation was avoided by the first enriching for the high temperature race in short time involved, not enough time for physi- culture medium at over 60 C followed by cloning ological acclimation was allowed. Maximal 14C with agar shake or streak methods at 45 C, no assimilation occurred at the temperature of col- cultures were produced that were sensitive to the 498 CASTENHOLZ BACTERIOL. REV. highest temperatures normally attained by this growth rate at 31 C was only about 20% of the organism [about 62 to 63 C (54)]. 45 to 53 C value (Fig. 4). Therefore, although Peary (145) found no deadaptation in one acclimation periods and lags differed, maximum clone of S. lividus after 8 months of growth at sustained growth rates could be attained at about 30 C, a distinctly sub-optimal temperature. At 1,200 ft-c for the entire temperature range that time it responded with the same growth rates mentioned. Photosynthetic saturation occurred from 55 to 30 C as the normal culture which had at about 800 ft-c in one Yellowstone strain of been maintained at 50 C. There were also no Synechoccus lividus at 45 C, when the cells were differences in lag periods, except at 30 C where the pregrown at that intensity (161). Similar values 30 C-grown material reached exponential phase often apply to Chlorella, but the preceding light first. The maintenance of the same clone of 0. intensity for growth has a great influence on the terebriformis at 29 to 30 and 44 to 45 C for 3 saturation intensity for photosynthesis (167). years has resulted in initially better growth rates A. nidulans appears to have a similarly high at 30 C for the 30 C stocks than for the 45 C saturation intensity even when pregrown at much stocks (unpublished data). However, the 30 C lower levels (118, 137). In natural hot spring mats, stock was still able to grow well at 50 C. Similarly, evidence has already been presented that suggests Bunning and Herdtle (49) reported higher photo- that many cells making up the algal layer might synthetic rates at 20 to 30 C in 0. geminata that not become light saturated even at very high mid- had been grown in that range than for those day intensities because of self-shading. grown at the optimum of 40 C. The photosyn- The high light intensities reaching the algal thetic rates at 40 C were similar for both cultures. surface in many thermal waters should result in M. laminosus was collected (by T. D. Brock) from "sun-adapted" forms. Typical manifestations of 58 C water in Iceland in 1966 but was isolated and "sun adaptation" are the lack of inhibition by cultivated in my laboratory at 45 C for 2.5 years. bright light and saturation of photosynthesis and Upon raising the temperature to 62 C, the culture growth by a relatively high intensity (see refer- continued to grow and appeared similar to ences 44, 167). Some of Brock's field results indi- others cultured at 62 C (54). Another strain which cate that there is little or no inhibition of photo- had a lower initial maximum (about 57 C) re- synthesis in the compact mats of Synechococcus tained this characteristic. Therefore, although loss at the highest natural light intensities. My results of growth potential at the upper or lower end of with suspensions of Synechococcus cells demon- the temperature range may be real in some cases, strate that inhibition at the level of an individual it is not general. cell does occur, but probably only above about In some thermophilic blue-greens, the growth 5,000 ft-c [about 0.2 cal per cm2 per min (400 to temperature optimum is a broad plateau, but it is 700 nm)]. With the high intensity values for usually skewed toward the upper end of the full saturation and inhibition, S. lividus qualifies well temperature range (Fig. 4). In the case of 0. as a sun-adapted organism. In contrast, some terebriformis, the maximal growth rate was re- blue-green algae are inhibited by light intensities tained to about one degree from the normal lower than 500 ft-c (44). lethal temperature of 54 C. The curve for Masti- gocladus growth, which is broad and well graded Effects of Light and Temperature at both ends, was sharpened at the upper end with on Pigmentation CO2 increase (105). Except for thermophilic organisms that may In thermophilic blue-green algae, the growth avoid bright light by growing at some depth in the temperature optimum or range may be modified microbial mat or by responding phototactically by other external factors; light intensity has one and gliding away from it (53), most thermophilic of the most profound effects. Most of the curves microorganisms are probably able to tolerate in Fig. 4 reflect changes in temperature under the exposures by making physiological adjustments same light intensity. In 0. terebriformis, however, which usually involve changes in chlorophyll and growth rates were measured after full acclimation carotenoid pigments. at each temperature under a light intensity Some cyanophytes, at least, seem to contain saturating for that temperature. An air (CO2) carotenoids which are not efficiently coupled to supply sufficient to saturate growth for each light photosynthesis (70, 93, 111, 140). Goedheer (93) intensity and temperature was also used. In this has evidence to suggest that energy transfer from species, the growth-saturating light intensity a-carotene to chlorophyll occurs in photosystem varied from about 1,200 ft-c (Coolwhite fluo- I in blue-green algae and that light energy ab- rescence) at 45 C to about 350 ft-c at 31 C, and sorbed by xanthophylls is not transferred to inhibitory effects of bright light were felt at about chlorophyll. Some of the carotenoids in most (or 5,000 and 1,500 ft-c, respectively. The maximum all) microorganisms function in protecting living VoL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 499 cell constituents from photooxidations sensitized ments as light intensities are increased should be by chlorophyll and other pigments (81, 119). This expected on the grounds that the excess pigment has been demonstrated in photosynthetic bacteria would not be required. A decrease in the amount (163), heterotrophic bacteria (135), and eucary- of photochemical machinery during bright-light otic photoautotrophs (see references 119, 138). exposure and the maintenance of relatively large However, the physicochemical nature of the amounts of carotenoids at the thylakoids should molecular interactions which constitute protec- be ofvalue in reducing photodynamic lesions. tion are still poorly understood (81). At a constant temperature and light intensity it At this point, there is no reason to suggest that was characteristic of the chlorophyll and carot- there is any consistent difference in the pigments enoid content of cells (pigment per unit of dry of thermophilic and nonthermophilic blue-green weight) to remain fairly constant during the algae (however, see reference 22). Phycoerythrin exponential phase of growth in 0. terebriformis, is absent from S. lividus, M. laminosus, and with self-shading kept to a minimum (unpublished apparently all blue-green algae which grow at data). At 45 C, the mean chlorophyll and carot- temperatures above 60 C. It is also lacking in the enoid values for the exponential growth phase mesophiles A. nidulans, Gloeocapsa alpicola, and varied inversely with the light intensity. From 200 P. luridwn (44). Phycoerythrin occurs in addition to 4,000 ft-c, there was approximately an eight- to phycocyanin in thermophilic S. minervae fold decrease in the characteristic amounts of (maximum, 60 C), 0. terebriformis "thermal- chlorophyll per cell during exponential growth, red" (maximum, 53 C), and in numerous meso- but the decrease in total carotenoids was only philes (unpublished data). The identity of some of about twofold. The ratio of carotenoid to chloro- the carotenoids in blue-green algae is still un- phyll (optical density at 472 nm to optical density certain and there is variation among species at 665 nm, in methanol) was about 0.5 to 0.7 (97-100). Myxoxanthophyll, echinenone, and j3- when grown in dim light in contrast to about 2.5 carotene, however, appear to be almost uni- when grown at 4,000 ft-c and higher. The de- versally present. It has been known for many years crease in chlorophyll with ascending light in- that the amount of chlorophyll decreases and tensities was accompanied by a large decrease in that the carotenoid content increases (at least phycoerythrin. Thus, 0. terebriformis t-r (ther- relative to chlorophyll) when blue-green algae are mal-red) was a deep red-brown color at low light grown under bright light. Complementary chro- intensity, due primarily to the large amounts of matic adaptations (87, 112) have also been phycoerythrin, whereas at very high intensity it described in blue-green algae, but these are was a pale ochre-yellow with carotenoids pre- probably over ridden by high intensity responses dominant. Growth of 0. terebriformis in turbu- in most hot-spring environments. lent aerated cultures at high light intensities re- Sargent (150) found that high intensity white sulted in approximately 100 times more myxo- light or light of various broad spectral bands xanthophyll than at the low intensities, although caused a change in Gloeocapsa montana from dark the other carotenoids were reduced. blue-green to yellow, and that lowering the in- At optimal temperatures (45 to 53 C), abrupt tensity reversed the process. The color changes upward or downward shifts in light intensity could be explained solely by a chlorophyll de- (involving a range of up to 4,000 ft-c) caused no crease over a few cell generations. A. nidulans distress in the cultures. An upward shift to bright (137) and Phonnidium persicinum (44) responded light resulted in an immediate increase in growth similarly. In Anacystis, phycocyanin paralleled rate (dry weight increase). Carotenoid and chloro- chlorophyll in its decline with increasing light in- phyll pigments adjusted to new levels and new tensity (no phycoerythrin), and in P. persicinum ratios in about 12 to 24 hr after the shift. There- phycoerythrin was even more sensitive to bright after they generally increased at the same rate light than chlorophyll (no phycocyanin present). as total dry weight (unpublished data). In the field, the greening and yellowing of thermal Synechococcus populations is controlled by the Growth and Survival at Suboptimal amount of shading, at least in summer (41, 52). In Temperaures S. lividus (41) and A. nidulans (8) grown at high light intensities, the decrease in chlorophyll was When 0. terebriformis cultures were shifted apparently accompanied by a decrease in the from an optimal to a suboptimal temperature, photosynthetic thylakoids. In the light intensity the initial pigment content appeared to be a range for saturation, photosynthesis is no longer critical quantity (unpublished data). If the cultures limited by the amount of light received but by the were preconditioned by growth at 45 C at a high maximum rate of the "dark" reactions. A con- light intensity (e.g., >2,000 ft-c) there was no lag tinuing decline in the bulk, light-harvesting pig- in growth when the culture was shifted to 31 C 500 CASTENHOLZ BACTERIOL. REV. at 350, 860, or 1,500 ft-c, but the rate dropped to tion of essential cellular (thylakoidal?) com- that typical of light saturation at 31 C (Fig. 4). ponents which include chlorophyll, the sensitizing However, if cells were preconditioned at a low pigment. In the absence of syntheses, or with very light intensity (350 ft-c) at 45 C and then shifted slow synthetic rates, these photodynamic lesions to 800 or 1,500 ft-c at 31 C, a lag of 6 to 7 days would eventually cause the death of the cell and occurred, followed by a gradual attainment of its subsequent lysis. Even when initially high exponential growth. The preconditioning under carotenoid levels apparently afford complete dim light at 45 C resulted in low carotenoid- photoprotection at first, eventually dispropor- chlorophyll ratios (about 0.5); in bright light, tionate changes in pigments will occur at tempera- values of 1.0-2.5 were obtained. In the lagging tures where synthetic rates are too slow to main- cultures, exponential decreases in chlorophyll tain balances. occurred for the period of the lag. The carotenoid In nature, it would seem that 0. terebriformis drop was not as great proportionately. Only after (and perhaps other blue-green algae, too) would high pigment ratios were finally attained at the lack the resilience to withstand abrupt increases lower temperature did exponential growth begin. in light intensity when stranded at suboptimal A shift to 28 C and 300 ft-c (after growing at 45 C temperatures, unless pigment balances were and 200 ft-c) resulted in an initial increase in dry favorable when the temperature change occurred. weight for a period of about 48 hr while chloro- Analyses of the pigments of natural populations phyll was simultaneously declining. This was of 0. terebriformis at various seasons have followed by a decrease in dry weight until about shown, however, that carotenoid-chlorophyll the sixth day when almost all of the cells had ratios are very much lower than for comparable lysed. An initial increase as described was not light intensity regimes in the laboratory (un- uncommon in downward shifts of temperature, published data). Nevertheless, this should be even to temperatures below the growth minimum. expected, since this and other species of blue- Cultures grown in bright light (about 5,000 ft-c), green algae in the springs occur as dense mats however, withstood the shift to 28 C without a with much self-shading. In the case of the Oscil- lag and grew at slow exponential rates for several latoria mats, the trichomes are so motile that no days. Even with properly preconditioned cells, individual is likely to maintain an exposed posi- 28 C seemed to be about the lowest temperature tion for very long. Within an optimal temperature at which aerated cultures could be maintained. range, great increases in light intensity should be Cells grown at 45 C and 5,000 ft-c were also tolerated even by the most exposed trichomes. shifted to 27 and 25 C. These grew for a limited However, it is probable that this resilience would time at very low rates (0.2 to 0.4 doublings per be lost if a trichome were washed downstream to day at 27 C), but eventually diminished to zero. lower temperatures or if the thermal gradient The highest growth rates of separate cultures steepened suddenly. Continued growth or sur- differed from each other under identical condi- vival would probably depend in large part on the tions at 27 or 25 C, apparently reflecting differ- pigment content of the cells at the time. ences in initial pigment content. The probable protection provided by carot- ACKNOWLEDGMMENTS enoids against detrimental lesions was also I gratefully acknowledge aid from grant GB-7103 from the demonstrated in light at much lower temperatures National Science Foundation for some of the work reported here. with 0. terebriformis. At 13 or 18 C, the death rate may be reduced considerably if cells have an LITERATURE CITED initially high carotenoid to chlorophyll ratio. At 1. Aasen, A., and S. L. Jensen. 1966. The carotenoids of Flexi- bacteria. III. The structures of flexixanthin and deoxy- 700 ft-c, the lysis of almost the entire cell popula- flexixanthin. Acta Chem. Scand. 20:1970-1988. tion occurred in 24 hr or less at 13 or 18 C when 2. Allen, E. T., and A. L. Day. 1935. Hot springs of the Yellow- the initial pigment ratios were below 1.0. With stone National Park. Carnegie Inst. Wash. Publ. 466. ratios 2.5), no significant lysis 3. Allen, M. B. 1950. The dynamic nature of thermophily. J. higher (about Gen. Physiol. 33:205-214. occurred for about 3 to 4 days, and during the 4. Allen, M. B. 1952. The cultivation of Myxophyceae. Arch. first 24 hr there was even a slight increase in dry Mikrobiol. 17:34-53. weight. The death rate in darkness at similar 5. Allen, M. B. 1953. The thermophilic aerobic sporeforming was ofpigment content, bacteria. Bacteriol. Rev. 17:125-173. temperatures independent 6. Allen, M. B. 1959. Studies with Cyanidium caldarium, an and there were large populations of viable cells anomalously pigmented chlorophyte. Arch. Mikrobiol. after 2 weeks at temperatures ranging from 10 to 32:270-277. 25 C. 7. Allen, M. B., and D. I. Amon. 1955. Studies on nitrogen- These results indicate that the most deleterious fixing blue-green algae. I. Growth and nitrogen-fixation by Anabaena cylindrica Lemm. Plant Physiol. 30:366-372. effect of exposure to suboptimal or subminimal 8. Allen, M. M. 1968. Photosynthetic membrane system in temperatures in the light may be the photooxida- Anacystis nidulans. J. Bacteriol. 96:836-841. VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 501

9. Allen, M. M. 1968. Ultrastructure of the cell wall and cell 35. Brock, T. D., and M. L. Brock. 1966. Temperature optima division of unicellular blue-green algae. J. Bacteriol. for algal development in Yellowstone and Iceland hot 96:842-852. springs. Nature 209:733-734. 10. Allen, M. M. 1968. Simple conditions for growth of unicellu- 36. Brock, T. D., and M. L. Brock. 1967. The measurement of lar blue-green algae on plates. J. Phycol. 4:1-4. chlorophyll, primary productivity, photophosphorylation, 11. Allen, M. M., and R. Y. Stanier. 1968. Selective isolation of and macromolecules in benthic algal mats. Limnol. blue-green algae from water and soil. J. Gen. Microbiol. Oceanogr. 12:600-605. 51:203-209. 37. Brock, T. D., and M. L. Brock. 1967. The hot springs of the 12. Amelunxen, R. E. 1966. Crystallization of thermostable Furnas Valley, Azores. Inter. Rev. Ges. Hydrobiol. 52: glyceraldehyde-3-phosphate dehydrogenase from Bacillus 545-558. stearothermophilus. Biochim. Biophys. Acta 122:175-181. 38. Brock, T. D., and M. L. Brock. 1968. Measurement of steady- 13. Amsbury, D. L. 1968. Terrestrial volcanic belts. Science state growth rates of a thermophilic alga directly in nature. 161:298. J. Bacteriol. 95:811-815. 14. Anagnostidis, K. 1961. Untersuchungen Uber die Cyano. 39. Brock, T. D., and M. L. Brock. 1969. Recovery of a hot phyceen einiger Thermen in Griechenland (In Greek, spring community from a catastrophe. J. Phycol. 5:75-77. German summary and discussion). Inst. Syst. Bot. Pflan- 40. Brock, T. D., and M. L. Brock. 1969. The fate in nature of zengeogr., University of Thessaloniki. photosynthetically assimilated 14C in a blue-green alga. 15. Anderson, G. C. 1958. Some limnological features of a Limnol. Oceanogr. 14:604-607. shallow saline meromictic lake. Limnol. Oceanogr. 3:259- 41. Brock, T. D., and M. L. Brock. 1969. Effect of light intensity 270. on photosynthesis by thermal algae adapted to natural 16. Ascione, R., W. Southwick, and J. R. Fresco. 1966. Labora- and reduced sunlight. Limnol. Oceanogr. 14:334-341. tory culturing of a thermophilic alga at high temperature. 42. Brock, T. D., and H. Freeze. 1969. Thermus aquaticus gen. Science 153:752-755. n. and sp. n., a nonsporulating extreme thermophile. J. 17. Barth, T. F. W. 1950. Volcanic geology, hot springs, and Bacteriol. 98:289-297. geysers of Iceland. Carnegie Inst. Wash. Publ. 587. 43. Brown, A. H., and G. C. Webster. 1953. The influence of 18. Bassel, A., and L. L. Campbell. 1969. Surface structure of light on the respiration of the blue-green alga Anabaena. Bacillus stearothermophilus ribosomes. J. Bacteriol. 98: Amer. J. Bot. 40:753-758. 811-815. 44. Brown, T. E., and F. L. Richardson. 1968. The effect of 19. Batterton, J. C., and C. Van Baalen. 1968. Phosphorus growth environment on the physiology of algae: light deficiency and the phosphate uptake in the blue-green alga intensity. J. Phycol. 4:38-54. .4nacystis nidulans. Can. J. Microbiol. 14:341-348. 45. Brues, C. T. 1928. Studies of the fauna of hot springs in the 20. Bauman, A. J., and P. G. Simmonds. 1969. Fatty acids and western United States and the biology of thermophilous polar lipids of extremely thermophilic filamentous bac- animals. Proc. Amer. Acad. Arts Sci. 63:139-228. terial masses from two Yellowstone hot springs. J. Bac- 46. Brues, C. T. 1932. Further studies on the fauna of North teriol. 98:528-531. American hot springs. Proc. Amer. Acad. Arts Sci. 67: 21. Bausum, H. T., and T. S. Matney. 1965. Boundary between 186-303. bacterial mesophilism and thermophilism. J. Bacteriol. 47. Bubela, B. 1968. Effect of temperature on growth character- 90:50-53. istics of Bacillus stearothermophilus. Aust. J. Biol. Sci. 22. Berns, D. S., and E. Scott. 1966. Protein aggregation in a 21:439-445. thermophilic protein. Phycocyanin from Synechococcus 48. Bubela, B., and E. S. Holdsworth. 1966. Protein synthesis in lividus. Biochemistry 5:1528-1533. Bacillus stearothermophilus. Biochim. Biophys. Acta 123: 23. Biebl, R., and E. Kusel-Fetzmann. 1966. Beobachtungen 376-389. liber das Vorkommen von Algen an Thermalstandorten 49. Bunning, E., and H. Herdtle. 1946. Physiologische Unter- auf Island. Osterreich. Bot. Zeit. 113:408-423. suchungen an thermophilen Blaualgen. Z. Naturforsch.

24. Biggins, J. 1969. Respiration in blue-green algae. 3. Bacteriol. 1:93-99. 99:570-575. 50. Card, G. L., C. E. Georgi, and W. E. Militzer. 1969. Phos- 25. Bodman, H., and N. E. Welker. 1969. Isolation of sphero- pholipids from Bacillus stearothermophilus. J. Bacteriol. plast membranes and stability of spheroplasts of Bacillus 97:186-192. stearothermophilus. J. Bacteriol. 97:924-935. 51. Castenholz, R. W. 1967. Aggregation in a thermophilic 26. Bott, T. L., and T. D. Brock. 1969. Bacterial growth rates Oscillatoria. Nature 215:1285-1286. at temperatures above 90 C in Yellowstone hot springs. 52. Castenholz, R. W. 1967. Environmental requirements of Science 164:1411-1412. thermophilic blue-green algae, p. 55-79. In A. F. Bartsch 27. Boye Petersen, J. 1923. The fresh-water Cyanophyceae of (ed.), Environmental requirements of blue-green algae. Iceland, p. 251-324. In The botany of Iceland. Vol. II. Federal Water Pollution Control Administration, Corval- Copenhagen. lis, Ore. 28. Brock, M. L., R. G. Wiegert, and T. D. Brock. 1969. Feeding 53. Castenholz, R. W. 1968. The behavior of Oscillatoria terebri- by Paracoenia and Ephydra (Diptera: Ephydridae) on the formis in hot springs. J. Phycol. 4:132-139. microorganisms of hot springs. Ecology 50:192-200. 54. Castenholz, R. W. 1969. The thermophilic cyanophytes of 29. Brock, T. D. 1967. Life at high temperatures. Science 158: Iceland and the upper temperature limit. J. Phycol. 5: 1012-1019. 350-358. 30. Brock, T. D. 1967. Relationship between standing crop and 55. Castenholz, R. W. 1970. Laboratory culture of thermophilic primary productivity along a hot spring thermal gradient. cyanophytes. Schweiz. Z. Hydrol., in press. Ecology 48:566-571. 56. Chapman, D. 1967. The effect of heat on membranes and 31. Brock, T. D. 1967. Micro-organisms adapted to high tem- membrane constituents, p. 123-146. In A. H. Rose (ed.), peratures. Nature 214:882-885. Thermobiology. Academic Press Inc., New York. 32. Brock, T. D. 1968. Taxonomic confusion concerning certain 57. Claus, G. 1959. Studien uber die Algenvegetation der Ther- filamentous blue-green algae. J. Phycol. 4:178-179. malquelle von Bilkkzhk, Nordungarn. Arch. Hydrobiol. 33. Brock, T. D. Vertical zonation in hot spring algal mats. 55:1-29. Phycologia, in press. 58. Cooney, D. G. and R. Emerson. 1964. Thermophilic fungi. 34. Brock, T. D. 1969. Microbial growth under extreme condi- W. H. Freeman, San Francisco. tions, p. 15-41. In P. Meadow and S. J. Pirt (ed.), Micro- 59. Copeland, J. J. 1936. Yellowstone thermal Myxophyceae. bial growth, Symposia of the Society for General Micro- Ann. N.Y. Acad. Sci. 36:1-229. biology XIX. 60. Craig, L. W., C. K Leach, and N. G. Carr. 1969. Studies with 502 CASTENHOLZ BACrERIOL. REV.

deoxyribonucleic acid from blue-green algae. Arch. Mikro- 89. Gates, D. M. 1962. Energy exchange in the biosphere. biol. 65:218-227. Harper & Row, New York. 61. Day, A. L., and E. T. Allen. 1925. The volcanic activity and 90. Geitler, L. 1932. Cyanophyceae. In L. Rabenhorst's Krypto- hot springs of Lassen Peak. Carnegie Inst. Wash. Publ. gamenflora von Deutschland, Osterreich und der Schweiz, 360. vol. 14. 62. Dor, I. 1967. Algues des sources thermales de Tiberiade. Sea 91. Geitler, L., and F. Ruttner. 1935. Die Cyanophyceen der Fish Res. Sta. Haifa Bull. 48:3-29. Deutschen Limnologischen Sunda-Expedition, ihre 63. Drouet, F. 1963. Ecophenes of Schizothrix calcicola. Proc. Morphologie, Systematik und (Okologie. Arch. Hydrobiol. Acad. Natur. Sci. Philadelphia 115:261-281. Suppl. 14:308-483; 1936, Suppl. 14:553-715. 64. Drouet, F. 1968. Revision of the classification of the Oscilla- 92. Gerloff, G. C. 1968. The comparative boson nutrition of toriaceae. Acad. Nat. Sci. Philadelphia Monogr. 15. several green and blue-green algae. Physiol. Plant. 21:369- 65. Drouet, F., and W. A. Daily. 1956. Revision of the coccoid 377. Myxophyceae. Butler Univ. Bot. Stud. 12:1-218. 93. Goedheer, J. C. 1969. Energy transfer from carotenoids to 66. Dyer, D. L., and R. D. Gafford. 1961. Some characteristics chlorophyll in blue-green, red and green algae and green- of a thermophilic blue-green alga. Science 134:616-617. ing bean leaves. Biochim. Biophys. Acta 172:252-265. 67. Dyer, D. L., and R. D. Gafford. 1963. The use of Synechococ- 94. Goldman, C. R. 1965. Micronutrient limiting factors and cus lividus in photogas exchangers, p. 87-107. In Develop- their detection in natural phytoplankton populations. ments in industrial microbiology, vol. 3. Mem. Ist. Ital. Idrobiol. 18 (Suppl):121-135. 68. Edelman, M., D. Swinton, J. A. Schiff, H. T. Epstein, and 95. Goldman, C. R., D. T. Mason, and J. E. Hobbie. 1967. Two B. Zeldin. 1967. Deoxyribonucleic acid of the blue-green Antarctic desert lakes. Limnol. Oceanogr. 12:295-310. algae (Cyanophyta). Bacteriol. Rev. 31:315-331. 96. Harrington, D., and J. C. Wright. 1966. Primary production 69. Edwards, M. R., D. S. Berns, W. C. Ghiorse, and S. C. Holt. in the Firehold River, Yellowstone National Park, Wyom- 1968. Ultrastructure of the thermophilic blue-green alga, ing (Abstr.). Amer. Soc. Limnol. Oceanogr. 29th Annual Synechococcus lividus Copeland. J. Phycol. 4:283-298. Meeting, p. 12. 70. Emerson, R., and C. M. Lewis. 1942. The photosynthetic 97. Healey, F. P. 1968. The carotenoids of four blue-green algae. efficiency of phycocyanin in Chroococcus, and the problem J. Phycol. 4:126-129. of carotenoid participation in photosynthesis. J. Gen. 98. Hertzberg, S., and S. L. Jensen. 1966. The carotenoids of Physiol. 25:579-595. blue-green algae. I. The carotenoids of Oscillatoria rube- 71. Emoto, Y. 1962. A bibliography of the thermal flora of scens and an Arthrospira sp. Phytochemistry. 5:557-563. Japan. J. Jap. Bot. 37:89-94, 119-124, 129-138. 99. Hertzberg, S., and S. L. Jensen. 1966. The carotenoids of 72. Epstein, I., and N. Grossowicz. 1969. Intracellular protein blue-green algae. II. The carotenoids of Aphanizomenon breakdown in a thermophile. J. Bacteriol. 99:418-421. flos-aquae. Phytochemistry 5:565-570. 73. Famin, M. A. 1933. Action de la temperature sur les vigd- 100. Hertzberg, S., and S. L. Jensen. 1967. The carotenoids of taux. Rev. Gen. Botan. 45:574-595, 655-682. blue-green algae. IIL. A comparative study of mutato- 74. Farrell, J., and A. H. Rose. 1967. Temperature effects on chrome and flavacin. Phytochemistry 6:1119-1126. microorganisms, p. 147-218. In A. H. Rose (ed.), Thermo- 101. Hoare, D. S., S. L. Hoare, and R. B. Moore. 1967. The biology. Academic Press Inc., New York. photoassimilation of organic compounds by autotrophic 75. Fay, P. 1965. Heterotrophy and nitrogen fixation in Chloro- blue-green algae. J. Gen. Microbiol. 49:351-370. gloea fritschil. J. Gen. Microbiol. 39:11-20. 102. Holm-Hansen, 0. 1963. Viability of blue-green and green 76. Fay, P., W. D. P. Stewart, A. E. Walsby, and G. E. Fogg. algae after freezing. Physiol. Plant. 16:530-540. 1968. Is the heterocyst the site of nitrogen fixation in blue- 103. Holm-Hansen, 0. 1964. Viability of lyophilized algae. Can. green algae? Nature 220:810-812. J. Bot. 42:127-137. 77. Festenstein, G. N., J. Lacey, F. A. Skinner. P. A. Jenkins, 104. Holm-Hansen, 0. 1968. Ecology, physiology, and biochemis- and J. Pepys. 1965. Self-heating of hay and grain in Dewar try of blue-green algae. Annu. Rev. Microbiol. 22:47-70. flasks and the development of farmer's lung antigens. J. 105. Holton, R. W. 1962. Isolation, growth, and respiration of a Gen. Microbiol. 41:389-407. thermophilic blue-green alga. Amer. J. Bot. 49:1-6. 78. Fogg, G. E. 1951. Studies on nitrogen fixation by blue-green 106. Hoogenhout, H., and J. Amesz. 1965. Growth rates ofphoto- algae. II. Nitrogen fixation by Mastigocladus laminosus synthetic micro-organisms in laboratory cultures. Arch. Cohn. J. Exp. Bot. 2:117-120. Mikrobiol. 50:10-25. 79. Fogg, G. E. 1956. The comparative physiology and biochem- 107. Horton, A. A. 1968. NADH oxidase in blue-green algae. istry of the blue-green algae. Bacteriol. Rev. 20:148-165. Biochem. Biophys. Res. Commun. 32:839. 80. Fogg, G. E. 1969. The physiology of an algal nuisance. Proc. 108. Hughes, E. O., P. R. Gorham, and A. Zehnder. 1958. Toxic- Roy. Soc. Ser. B Biol. Sci. 173:175-189. ity of a unialgal culture of Microcystis aeruginosa. Can. J. 81. Foote, C. S. 1968. Mechanisms of photosensitized oxidation. Microbiol. 4:225-236. Science 162:963-969. 109. Jaag, 0. 1945. Untersuchungen uiber die Vegetation und 82. Forsberg, C. 1965. Nutritional studies of Chara in axenic Biologie der Algen des nackten Gesteins in den Alpen, im culture. Physiol. Plant. 18:275-290. Jura und im schweizerischen Mittelland. Beitr. Krypto- 83. Forsberg, C., and G. Lindqvist. 1967. On biological degrada- gamenflora Schweiz 9 (3):1-560. tion of nitrilotriacetate (NTA). Life Sci. 6:1961-1962. 110. Jitts, H. R., C. D. McAllister, K. Stephens, and J. D. H. 84. Fox, D. L., and R. A. Lewin. 1963. A preliminary study of Strickland. 1964. The cell division rates of some marine the carotenoids of some flexibacteria. Can. J. Microbiol. phytoplankton as a function of light and temperature. J. 9:753-768. Fish. Res. Board Can. 21:139-157. 85. Fremy, P. 1936. Remarques sur la morphologie et la biologie 111. Jones, L. W., and J. Myers. 1964. Enhancement in the blue- 'e l'Hapalosiphon laminosus Hansg. Ann. Protistologie green alga, Anacystis nidulans. Plant Physiol. 39:938-946. 5:175-200. 112. Jones, L. W., and J. Myers. 1965. Pigment variation in 86. Friedman, S. M. 1968. Protein-synthesizing machinery of Anacystis nidulans induced by light of selected wavelengths. thermophilic bacteria. Bacteriol. Rev. 32:27-38. J. Phycol. 1:7-14. 87. Fujita, Y., and A. Hattori. 1960. Effect of chromatic lights 113. Kempner, E. S. 1963. Upper temperature limit of life. Science on phycobilin formation in a blue-green alga Tolypoihrlx 142:1318-1319. tenuis. Plant Cell Physiol. 1:293-303. 114. Kiyohara, T., Y. Fujita, A. Hattori, and A. Watanabe. 1960. 88. Gates, D. M. 1961. Winter thermal radiation studies in Heterotrophic culture of a blue-green alga, Tolypothrix Yellowstone Park. Science 134:32-35. tenuis. J. Gen. Appl. Microbiol. 6:176-182. VOL. 33, 1969 THERMOPHILIC BLUE-GREEN ALGAE 503

115. Klas, S., and E. Marnenko. 1959. Mikrovegetacija termalnog 141. O'Donovan, G. A., and J. L. Ingraham. 1965. Cold-sensitive vrela Sv. Helena Kod Samobora. Jugo. Akad. 7nanosti mutants of Escherichia coli resulting from increased feed Umjetnosti '"Rad" 317:243-290. back inhibition. Proc. Nat. Acad. Sci. U.S.A. 54:451-457. 116. Kol, E. 1932. 'Ober die Algenvegetation der Hajddszoboszl6er 142. O'Kelly, J. C. 1968. Mineral nutrition of algae. Annu. Rev. Therme. Arch. Protisten. 76:309-324. Plant Physiol. 19:89-112. 117. Kratz, W. A., and J. Myers. 1955. Nutrition and growth of 143. Pearce, J., and N. G. Carr. 1967. The metabolism of acetate several blue-green algae. Amer. J. Bot. 42:282-287. by the blue-green algae, Anabaena varlabilis and Anacystis 118. Kratz, W. A., and J. Myers. 1955. Photosynthesis and respi- nidulans. J. Gen. Microbiol. 49:301-313. ration of three blue-green algae. Plant Physiol. 30:275-280. 144. Pearce, J., C. K. Leach, and N. G. Carr. 1969. The incom- 119. Krinsky, N. I. 1966. The role of carotenoid pigments as plete tricarboxylic acid cycle in the blue-green alga Ana- protective agents against photosensitized oxidations in baena variabills. J. Gen. Microbiol. 55:371-378. chloroplasts, p. 423-430. In T. W. Goodwin (ed.), Bio- 145. Peary, J. 1964. Ecology and growth studies of thermophilic chemistry of chloroplasts, vol. I. Academic Press Inc., blue-green algae. Ph.D. Thesis, University of Oregon, New York. Eugene. 120. Kullberg, R. G. 1968. Algal diversity in several thermal spring 146. Peary, J., and R. W. Castenholz. 1964. Temperature strains effluents. Ecology 49:751-755. of a thermophilic blue-green alga. Nature 202:720-721. 121. Lang, N. J. 1968. The fine structure of blue-green algae. 147. Prmt, S. 1956. Zur Physiologie der Mineral-und Thermalwa- Annu. Rev. Microbiol. 22:15-46. sservegetation. Hydrobiologia 8:328-364. 122. Lazaroff, N. 1966. Photoinduction and photoreversal of the 148. Proctor, V. W. 1959. Dispersal of fresh-water algae by migra- Nostocacean developmental cycle. J. Phycol. 2:7-17. tory water birds. Science 130:623-624. 123. Leach, C. K. and H. G. Carr. 1969. Oxidative phosphoryla- 149. Proctor, V. W., C. R. Malone, and V. L. DeVlaming. 1967. tion in an extract of Anabaena varlabills. Biochem. J. Dispersal of aquatic organisms: viability of disseminules 112:125-126. recovered from the intestinal tract of captive Killdeer. 124. Lederberg, J. 1954. A simple method for isolating individual Ecology 48:672-676. microbes. J. Bacteriol. 68:258-259. 150. Sargent, M. C. 1934. Causes of color change in blue-green 125. Lewin, R. A. 1959. The isolation of algae. Rev. Algol. 4:181- algae. Proc. Nat. Acad. Sci. U.S.A. 20.251-254. 197. 151. Schopf, J. W. 1968. Microflora of the Bitter Springs Forma- 126. Lewin, R. A. 1965. Freshwater species of Saprospira. Can. J. tion, Late Precambrian, Central Australia. J. Paleontol. Microbiol. 11:135-139. 42:651-688. 127. Livingstone, D. A. 1963. Chemical composition of rivers 152. Schneider, K. C., C. Bradbeer, R. N. Singh, L. C. Wang, and lakes. U.S. Geological Survey Professional Paper P. W. Wilson, and R. H. Burris. 1960. Nitrogen fixation 440-G. by cell-free preparations from microorganisms. Proc. 128. Ldwenstein, A. 1903. Uber die Temperaturgrenzen des Natl. Acad. Sci., U.S.A. 46:726-733. Lebens bei der Thermalge Mastigocladus laminosus Cohn. 153. Schwabe, G. H. 1936. Beitrage zur Kenntnis isliindischer Ber. Deut. Bot. Ges. 21:317-323. Thermalbiotope. Arch. Hydrobiol. 6 (Suppl.):161-352. 129. Maciasr, F. M. and R. W. Eppley. 1963. Development of 154. Schwabe, G. H. 1960. Uber den thermobionten Kosmopoli- EDTA media for the rapid growth of Chlanydomonas tan Mastigocladus laminosus Cohn. Blau-Algen und mundana. J. Protozool. 10:243-246. Lebensraum V. Schweiz. Z. Hydrol. 22:757-792. 130. Mandel, M. 1969. New approaches to bacterial taxonomy: 155. Schwabe, G. H. 1962. Lagerbildungen bei Hormogonalen, perspective and prospects. Annu. Rev. Microbiol. 23:239- p. 53-60. In Beitrage zur Physiologie und Morphologie der 274. Algen. Gustav Fischer, Stuttgart. 131. Mann, J. E., and H. E. Schlichting. 1967. Benthic algae of 156. Schwabe, G. H. 1964. Lagerbildungen hormogonaler Blaual- selected thermal springs in Yellowstone National Park. gen in thermalen und anderen extremen Biotopen. Verh. Trans. Amer. Micros. Soc. 86:2-9. Int. Verein. Limnol. 15:772-781. 132. Margenko, E. 1962. Licht- und Elektronmikroskopische 157. Schwabe, G. H. 1966. Okologischer Charakter und System Untersuchungen an der Thermalalge Mastigocladus der Cyanophyten. Verh. Int. Verein. Limnol. 16:1541- laminosus Cohn. Acta Bot. Croatica 21:47-74. 1548. 133. Marre, E. 1962. Temperature, p. 541-550. In R. A. Lewin 158. Setchell, W. A. 1903. The upper temperature limits of life. (ed.), Physiology and biochemistry of algae. Academic Science 17:934-937. Press Inc., New York. 159. Sharp, J. H. 1969. Blue-green algae from Bermuda waters: 134. Marsh, C. L., and D. H. Larsen. 1953. Characterization of ecologically selected variations of a single species. J. some thermophilic bacteria from the hot springs of Yellow- Phycol. 5:53-57. stone National Park. J. Bacteriol. 65:193-197. 160. Sharp, J. H. 1969. Blue-green algae and carbonates-Schizo- 135. Mathews, M. M. and W. R. Sistrom. 1959. Function of thrix calcicola and algal stromatolites from Bermuda. carotenoid pigments in non-photosynthetic bacteria. Limnol. Oceanogr. 14:568-578. Nature 184:1892-1893. 161. Sheridan, R. P. 1966. Photochemical and dark reduction of 136. McDonald, W. C. and T. S. Matney. 1965. Genetic transfer sulfate and thiosulfate to hydrogen sulfide in Synechococcus of the ability to grow at 55 C in Bacillus subtilis. J. Bac- lividus. Ph.D. Thesis, University of Oregon, Eugene. teriol. 85:218-220. 162. Singh, R. N., and H. N. Singh. 1964. Ultra-violet induced 137. Myers, J., and W. A. Kratz. 1955. Relations between pigment mutants of blue-green algae. I. Anabaena cycadeae Reinke. content and photosynthetic characteristics in a blue-green Arch. Mikrobiol. 48:109-117. alga. J. Gen. Physiol. 39:11-22. 163. Sistrom, W. R., M. Griffiths, and R. Y. Stanier. 1956. The 138. Nakayama, T. 0. M. 1962. Carotenoids, p. 409-420. In biology of a photosynthetic bacterium which lacks colored R. A. Lewin (ed.), Physiology and biochemistry of algae. carotenoids. J. Cell. Comp. Physiol. 48:473-515. Academic Press Inc., New York. 164. Sladeckova, A. 1969. Control of slimes and algae in cooling 139. Nash, A. 1938. The Cyanophyceae of the thermal regions of systems. Verh. Internat. Verein. Limnol. 17: in press. Yellowstone National Park, U.S.A., and of Rotorua and 165. Smith, A. J., J. London, and R. Y. Stanier. 1967. Biochemical Whakarewarewa, New Zealand, with some ecological basis of obligate autotrophy in blue-green algae and data. Ph.D. Thesis, University of Minnesota, Minneapolis. thiobacilli. J. Bacteriol. 94:972-983. 140. Nultsch, W., and G. Richter. 1963. Aktionsspektrum des 166. Sorokin, C. 1960. Kinetic studies of temperature effects on photosynthetischen 14CO-Einbaus von Phormidium the cellular level. Biochim. Biophys. Acta 38:197-204. uncinatum. Arch. Mikrobiol. 47:207-213. 167. Steeman Nielsen, E., and E. G. Jorgensen. 1968. The adapta- 504 CASTENHOLZ BACrERIOL. REV.

tion of plankton algae. I. General part. Physiol. Plant. 177. Van Baalen, C. 1967. Further observations on growth of 21:401-413. single cells of coccoid blue-green algae. J. Phycol. 3:154- 168. Stewart, W. D. P. 1968. Nitrogen input into aquatic eco- 157. systems, p. 53-72. In D. F. Jackson. (ed.), Algae, man, and 178. Vincent, E. R. 1967. A comparison of riffle insect populations the environment. Syracuse University Press, Syracuse, N.Y. in the Gibbon River above and below the geyser basins, 169. Stockner, J. G. 1967. Observations of thermophilic algal Yellowstone National Park. Limnol. Oceanogr. 12:18-26. communities in Mount Rainier and Yellowstone National 179. Vouk, V. 1950. Grundriss zu einer Balneobiologie der Parks. Limnol. Oceanogr. 12:13-17. Thermen. Verlag BirkhaIuser, Basel. 170. Stockner, J. G. 1968. Algal growth and primary productivity 180. Walsby, A. E. 1968. Mucilage secretion and the movements in a thermal stream. J. Fish. Res. Board Can. 25:2037- of blue-green algae. Protoplasma 65:223-238. 2058. 181. Waring, G. A. 1965. Thermal springs of the United States 171. Swain, F. M. 1969. Paleomicrobiology. Annu. Rev. Micro- and other countries of the world-a summary. U.S. biol. 23:455-472. Geological Survey Professional Paper 492. 172. Szeicz, G. 1966. Field measurements of energy in the 0.4-0.7 182. Webster, G. C., and A. W. Frenkel. 1952. Some respiration micron range, p. 41-51. In R. Bainbridge, G. C. Evans, characteristics of the blue-green alga, Anabaena. Plant and 0. Rackham (ed.), Light as an ecological factor. Physiol. 28:63-69. John Wiley & Sons, New York. 183. White, D. E., 173. Thomas, J., and E. A. Gonzalves. 1965. Thermal algae of J. D. Hem, and G. A. Waring. 1963. Chemical Western India. Hydrobiologia 25:(I) 330-340, (II) 340- composition of sub-surface waters. U.S. Geological 351; 26:(II) 21-28, (IV) 29-40, (V) 41-54, (VI) 55-65. Survey Professional Paper 440-F. 174. Trtiper, H. G. 1969. Bacterial sulfate reduction in the Red 184. Whitford, L. A., and G. J. Schumacher. 1964. Effect of a Sea hot brines, in press. In E. T. Degens and D. A. Ross current on respiration and mineral uptake in Spirogyra (ed.), Hot brines and recent heavy metal deposits in the and Oedogoniwn. Ecology 45:168-170. Red Sea. Springer-Verlag, New York. 185. Williams, R. B. 1964. Division rates of salt marsh diatoms in 175. Tuxen, S. L. 1944. The hot springs of Iceland, their animal relation to salinity and cell size. Ecology 45:877-881. communities and their zoogeographical significance, p. 186. Wyatt, J. T., and J. K. Silvey. 1969. Nitrogen fixation by 1-206. In The zoology of Iceland, vol. I, part 11. Einar Gloeocapsa. Science 165:908-909. Munksgaard, Copenhagen. 187. Yoneda, Y. 1952. A general consideration of the thermal 176. Uzamasa, Y. 1965. Chemical investigations of hot springs in Cyanophyceae of Japan. Memoirs College of Agriculture, Japan. Tsukiji Shokan, Tokyo. Kyoto University, Fisheries Ser. 62:1-20.