The Habitat of the Eurypterida/Chapter III

by Marjorie O'Connell

CHAPTER III

The Bionomy of the Eurypterid Faunas

INTRODUCTION

In the first chapter I confined myself to facts which consisted of observations made in the field or the laboratory by students of the rocks and of the faunas. Such facts covered data on the geological and geographical distribution of the eurypterids; in the second chapter I gave a resumé of the opinions which have been held by various writers in regard to the habitat of the eurypterids; the remainder of the paper will be devoted to the contemplation of the recorded facts whose interpretation will be undertaken in the light of principles recognized by the school of philosophical geologists. For this reason, I shall in nearly all cases use the deductive method of inquiry, establishing the general principles which may then be applied to the particular case in hand. It is evident, then, that before we can begin to adduce proofs favoring one mode of life or another for the eurypterids, we must have a good classification of habitats in which each type is clearly defined, and we must determine at the very outset whether there are any criteria which may be recognized in the rocks as absolutely diagnostic of the habitats of the past. In so far as deposits in the sea and on the land have received any consideration at all, distinctions have mainly been drawn on the physical character; of the sediments; but I believe that much more accurate and far reaching results are to be obtained from the study of the fossil faunas. These may be investigated from two points of view, either the chorological, or the bionomic; and besides these, there is yet a third line of approach, namely the geological, in which the physical characters and lithogenesis of the sediments, together with the correlation of synchronous deposits constitute the elements. These three lines of investigation deal with three, for the most part mutually independent groups of facts and I am convinced that any one will yield sufficient evidence to determine the nature of any past habitat. In this chapter I shall deal with the bionomic characteristics of modern habitats, and shall give the criteria for recognizing ancient ones, concluding with the special case of the bionomy of the eurypterids. The following chapter will be devoted to the geological evidence regarding the habitats, while I shall defer until the fifth chapter the chorological evidence which is more conveniently discussed with the geological occurrences in mind. In considering the habitats, we may confine ourselves to those which are aqueous, since certain anatomical features, such as the nature of the cephalothoracic appendages and the presence of branchiae on the abdominal appendages, establish beyond a doubt the fact that the eurypterids lived in the water and not on the land, as do their near relatives the scorpions.

Before attempting to draw conclusions about the conditions under which the Eurypterida must have lived, it is necessary to have in mind the physical and faunal characteristics of the various types of habitats. Since these characteristics have never, so far as the writer knows, been discussed at length, I shall here state some of the results of an extended study of aqueous habitats which in time will be published as a separate paper.^[1]

CLASSIFICATION OF RECENT AQUEOUS HABITATS

The most natural and fundamental characteristic which can be recognized in classifying aquatic bionomic realms is salinity, on which basis it is readily seen that there are only two original habitats: (1) marine, (2) terrestrial fresh water. Animals living either in marine or in fresh waters may become adapted to water which is of a salinity intermediate between the other two and generally designated "brackish," or to a salinity greater than that of normal marine waters. Thus to the two original types of habitat may be added two others: (3) brackish water, and (4) super-saline water, which are never original habitats. By this is meant that, minor variations excepted, no aquatic forms ever originate in the brackish water of estuaries, lagoons, cut-off arms of the sea, or interior basins, or in the super-saline waters of lakes. (This will be demonstrated below, pp. 73, 76, 77.) That this should be the case is due to the evanescent character of such water bodies. It is, of course, conceivable that a body of this type, long persistent, might be peopled from the land or from another aqueous realm and that such a fauna might be specialized. To these principal types may be added certain minor ones, giving seven in all. In the following table are given the salinity types, and with them what seems to be the best salinity ranges. It is not of importance here to go into the reasons for the making of the limits, but it may be said that they are based on a large number of typical examples in each case. The various realms may be secondarily grouped according to occurrence, as marine and terrestrial. In this latter group, two fur- ther subdivisions may be made according to mobility: the courant and the static waters. I propose the term courant for all terrestrial waters which are constantly moving in a given direction, as do the rivers. The following table brings out these points, of which further discussion will not be given in this paper.

CLASSIFICATION OF AQUEOUS BIONOMIC REALMS ACCORDING TO SALINITY^[2]

CLASSIFICATION		SELECTED EXAMPLES
Type of Salinity	Range in Permille	Marine		Courant		Static
		Examples	Per- mille	Examples	Per- mille	Examples	Per- mille
I. Fresh	0.0–0.2			Amazon River at Obidos	0.037	Lake Erie	0.134
				Rhine at Cologne	0.178
II. Subbrackish	0.2–1.0			Vistula near Culm	0.201	Laacher See	0.218
				Arkansas R.	0.794	Humboldt Lake	0.928
III. Brackish	1.0–10.0	Baltic Sea	7.80	Salt River	1.234	Palic Lake	2.215
				Rio de los Papagayoa	9.815	Lake Biljo	8.800
IV. Superbrackish	10.0–20.0	Sea of Azov	10.60			Caspian Sea (in 1878)	12.940
		Black Sea	18.30
V. Subsaline	20.0–30.0	Arctic Ocean	25.50			Van Lake	22.601
		Hudson Bay	26.00
VI. Saline	30.0–40.0	Behring Sea	30.30			Albert Lake	39.772
		Atlantic Ocean	35.37
		Red Sea	38.80
VII. Supersaline	40.0–289 plus					Tinetz Lake	289.000

RECENT AQUATIC FAUNAS

Having established what seem to be fairly accurate limits for the ranges in salinity in all of the waters on the surface of the earth, it becomes possible to study the faunas of these different realms, for the type of life in any given water body is more dependent upon the salinity than upon any other physical factor with the exception of extremes of temperature. The absolute necessity of studying recent faunas with particular attention to the types of organisms represented, and to the numbers of species and of individuals, has not been realized sufficiently in the past. The habitats of fossil faunas cannot be determined without a knowledge, an intimate knowledge, of the habitats of recent faunas. To be sure, there is little doubt about the kinds of organisms which make up a typical marine fauna; in many cases, too, there may be no difficulty in recognizing a fresh water (especially lake) fauna, but there is an undoubted haziness and lack of precision in all ideas connected with brackish waters and with the faunas thereof. When a given fossil fauna has shown certain peculiar characteristics, such, for instance, as a complete or almost complete absence of molluscan representatives or when the fauna has been confined to one or two classes of organisms, the custom has been and still is to say that the organisms lived in brackish water. It is, thus, necessary to determine the nature of recent faunas which are characteristic of the various bionomic realms, in order that we may, not without a fair degree of certainty, establish the criteria for determining the faunal nature of the habitats of the past.

Marine. The marine fauna is always large and varied, comprising, typically, representatives from each taxonomic division among the invertebrates. Not only are there a large number of genera and species, but nearly all phyla are represented. For the mollusca alone the number of genera in a given region may run up into the hundreds and that of the species may be considerably over a thousand. The figures apply especially to the littoral zone, that belt along all coasts which is most favorable to life. There light penetrates to the bottom, the food supply is abundant, and varying substrata are available to suit the needs of different organisms. This zone, extending from high water approximately to the two hundred fathom line, is the one of greatest geologic interest because nearly all of the marine formations of the past were littoral; unequivocal abyssal deposits being very rare. Since practically all of the invertebrate organisms of this prolific littoral marine fauna are protected either by shells or by exoskeletons, each individual that dies leaves its record behind in some hard part which falls to the bottom when the animal dies, or else soon comes to rest there, where it is buried by sand or mud. Not only are the remains of the animals which lived in the littoral zone of the sea preserved in the deposits forming there, but many derelicts, dead or alive, are washed in from the land and the rivers and we have a phenomenon observable in no other bionomic realm, namely, the commingling in one life district of the remains of organisms from all the other districts. During storms, terrestrial animals are drowned in the torrential floods, trees and other vegetation are carried away in the undermining of the banks, and these, together with the remains of fluviatile organisms and even with the living forms which cannot resist the strength of the current, are all carried out to sea to be dropped and there entombed with the remains of marine organisms. In tropical and semi-arid regions such mingling of terrestrial and marine forms is the common, not the unusual, thing. Darwin has called attention to many such cases in his Voyage of the Beagle, where he describes the great drought which occurred between the years 1827 and 1832 in Buenos Ayres, South America, when the birds and animals died by the thousand, the vegetation became withered and parched, and the dry winds swept over the desolate waste of land desiccated and dusty. The large rivers shrivelled, the small ones disappeared altogether; and where a little water still remained in the broader courses, it became highly saline, bringing death to the animals who drank. Herds of cattle rushed into the river, crazed by thirst, and there perished from the salt water and because they were too weak to climb up the banks again. Following this drought which lasted five years, came the rainy season and torrential floods. "Hence it is almost certain," Darwin concludes, "that some thousands of the skeletons were buried by the deposits of the very next year" (48, 127). Not only in semi-arid climates where torrential floods are active, but even in pluvial climates are terrestrial and fluviatile organisms carried out to the littoral zone of the sea, where they are buried in the delta deposits together with marine shells and tests. Thus, terrestrial vertebrate remains have been found in the deltas of the Ganges and Zambesi, the bones of recent antelope, buffalo, lion, hippopotamus and other mammals having been recorded; in the Po delta arthropods occur with lignites. Such terrestrial relics are by no means confined to deltas or river flood plains, but are found along all coasts even where no rivers enter as well as at considerable distances from shore beyond the debouchures.

Walther makes mention of the occurrence of great rafts of trees off the mouth of the Congo, 450 km. from shore, some of these interlocking tree islands being 100 m. across. Agassiz likewise has noted in the Caribbean Sea, Helix, leaves, and other land organisms dredged from a depth of 1000 to 2000 fathoms, which is far beyond the littoral zone.

Thus we must conclude that in the marine waters, and especially in the littoral zone, there is not only an abundance of invertebrate organisms of nearly all phyla, but there are stragglers from other realms; insects and plants are blown out to sea, while terrestrial animals and vegetal remains, together with fluviatile organisms, are carried along by the rivers, all at length being entombed in the marine sediments with the hard parts of the organisms which lived and died in the sea. In such cases we should expect to find the fluviatile and terrestrial remains shattered and worn on account of being transported oftentimes for a considerable distance, and usually subject to partial destruction by the débris which the rivers carry. At any rate, it is apparent that it is customary, not anomalous, for the remains of terrestrial and marine organisms to occur together.

Fresh Water. The fresh-water faunas of rivers and lakes, on the other hand, present quite different features. While the number of individuals in a given river or lake may indeed be large, the number of genera and species is very small as compared with those in the neighboring marine waters. Furthermore, there are only a few large classes abundantly represented, such as the fish, molluscs, and protozoa, while all of the other classes, so well represented in marine waters, are in given rivers or lakes represented often by a single species only, or by none at all. Three comparative sets of figures for the molluscs will serve to illustrate how small the number of genera and species is in fresh water when compared with those in marine waters at approximately the same latitude.

Table showing number of Genera and Species of Mollusca in Various Bionomic Realms

LOCALITY	NIAGARA RIVER	SAGINAW BAY	WOODS HOLE, MASS.	MASS. COAST
	permille	permille	permille	permille
Salinity	0.134	0.105	30.0	35.0
Genera	15	23	133	175
Species	24	93	203	466

The complete known invertebrate fauna from Woods Hole numbers 1286 species, while in the open marine, somewhat more saline waters, the number is even larger (270, 85).

Brackish Water. In dealing with the brackish-water faunas many difficulties are encountered, because not very much work has been done in connection with the various brackish-water bodies and it is thus hard to obtain data. Two examples, the Baltic Sea, and the Severn Estuary will be discussed.

The Baltic Sea. One of the best known of brackish-water bodies is the Baltic, which, though it cannot be considered an estuary may yet serve admirably to demonstrate the changes in fauna which occur with changes of salinity. The Baltic lacks the tides which are characteristic in estuaries and therefore does not exhibit the pronounced changes from fresh to salt water twice a day. It is more static and shows in a large way the responses of the fauna to salinity. The North Sea has a normal marine salinity of 35.00 permille, which decreases steadily eastward in the Baltic. In the Skager Rak it is 34 permille, off Skagen, the northeasternmost point of Denmark, it is 30 permille, in the Kattegat 22, and in the Bay of Kiel 20 permille. Throughout the southern part of the Baltic, from the "Scheren," at the mouth of the Gulf of Finland, to Bornholm the salinity is from 7 to 8 permille at the surface and does not vary greatly in the depths. For instance, in the deepest part of the Baltic off the Island of Gotland the salinity is only 12 permille, and in the Bay of Danzig, which shows a yearly average of 7.22 permille at the surface, it is only 11.66 permille (average) at the depth of 105 meters. In the Bay of Riga the salinity is 6 permille, in the southern part of the Gulf of Bothnia it is 4 permille and gradually diminishes until the water is entirely fresh. Corresponding to these changes in salinity are certain very definite changes in the fauna (Fig. 1).

As the salinity decreases from that of normal sea water, 35.00 permille, the fauna changes from a typical marine one to one in which only certain groups are represented and finally to an entirely fresh-water fauna. Each phylum shows this change; Pouchet and de Guerne have reported that a truly marine crustacean fauna extends into the Baltic as far as Kalmar Sound, between Öland and Sweden, but that beyond this point the marine species are gradually replaced by certain euryhaline forms and finally by the fresh-water ones until at the head of the Gulf of Finland the planktonic crustacean fauna is made up entirely of fresh-water types. Thus Evadne nordmanni is very abundant in the western part of the Baltic, but is replaced eastward by Bosmina longirostris; another abundant euryhaline form is Podon intermedius. The replacing fresh water types are such as Cyclops quadricornis, Daphnella brachyura, Daphnia quadrangula and

Fig. 1. Sketch Map or Baltic Showing Permillage Variations in Salinity

Bosmina longirostris (Pouchet and de Guerne, 223, 919, 920). The mollusca show a similar change, the Littorina species, for instance, being replaced by Limnæa, while along the coast where the salinity is low both of these forms live together, and there also Neritina fluviatilis, a river form is found.

A second change which takes place in the Baltic fauna and which may be correlated with the variation in salinity, is that the stenohaline forms of the marine fauna disappear altogether, while the euryhaline ones become dwarfed. Thus, the common cockle shell, Cardium edule, in the North Sea, of normal marine salinity, is the size of a small apple, at Stockholm, where the salinity is below 10 permille, the shell in the deeper, more saline water is only as large as a walnut and is even smaller along shore where the water is fresher. At Königsberg, with the decreasing salinity, the size reaches that of a hazel nut, whereas at Reval, it is only the size of a pea. In like manner, Mytilus edulis, which is 8 to 9 cm. long at Kiel is only 3 to 4 cm. long at Gotland. The fish and worms also show dwarfing.

A third point to be noted is that the fauna decreases very rapidly in the number of species which occur. Karl Möbius in his report on the faunal survey in the Baltic Sea made by the Pommerania, states that:

"The total number of observed invertebrate animals amounts to about 200 species, not including the infusoria and crinoids.

"We have found scarcely one-fifth of these in the great eastern basin of the Baltic which begins between Rügen and the southern extremity of Sweden" (182, 277). The following table shows the distribution for the invertebrates. The numbers given for the Baltic as a whole should be compared with those given for Kiel. In this latter bay the conditions are not so different from those in the open

Comparative Number of Species of Invertebrates in the Baltic, etc.

PHYLA	WATERS AROUND GREAT BRITAIN	BALTIC AS A WHOLE	BAY OF KIEL	BAY OF TRAVE- MUNDE
	35 permille	7.8 permille	20 permille	12 permille
Protozoa	69
Porifera	42	7	3	3
Coelenterata	98	28	24	8
Echinodermata	48	6	5	2
Vermes	101	68	50	26
Bryozoa		11	8	5
Crustacea		50	36	19
Mollusca	682	68	64	40
Tunicata		5	4	4
Total	1040 (+)	243	194	107

sea as to have affected the fauna very markedly, but it will be noted that marine types must predominate in the fauna of the Baltic, as a whole, since for none of the phyla are the numbers much greater than they are at Kiel, where the fauna is wholly marine.

Perhaps the most significant fact brought out, is that the marine forms which are found in the Baltic, though they may be dwarfed or otherwise modified, are not different specifically from the marine forms found along the coasts of Great Britain, nor do the fresh-water forms differ from those found in the rivers emptying into the Baltic, or those in the neighboring fresh-water bodies. Thus it is established that in a brackish-water body of the nature of the Baltic the fauna is due to the mingling of modified marine and of modified fresh-water, that is, river forms. Only the more euryhaline marine species survive and these may in a given estuary give rise to a fauna which we may designate as a "brackish-water fauna." It will consist of forms derived in the manner just described, and these forms may become adapted to the peculiar temperature and salinity conditions prevailing in the given estuary. Thus, new mutations, varieties, and occasionally species may arise, but seldom a new genus and never a whole class of organisms. It is only the smaller taxonomic divisions which are affected. Furthermore, a "brackish-water fauna" in any estuary is always ephemeral, for the estuary is of short duration, geologically speaking.

The Severn Estuary. While the Baltic serves to show on a large scale what happens to a marine fauna which is gradually subjected to fresher and fresher water until it passes through brackish conditions to entirely fresh ones, there is another type of brackish water, the estuary, which is often said to have a fauna of its own. An estuary may be defined as the drowned lower portion of a river in which twice daily there is a change in the water from fresh to marine and back again as the tide comes in and goes out. On account of the tidal scour and thorough mixing of the marine and the inflowing river water, the brackish portion will not be very large. The Severn, on the west coast of England, is a very typical estuary, having the long, slowly broadening form toward the sea. There are a number of tributaries with their respective estuaries, so that on the whole the Severn may be considered characteristic. It is well known that muds are the dominant sediments, not only in the main tidal channel far out to sea, but also in all of the tributary channels. Professor W. J. Sollas has made a careful study of these muds in order to determine their distribution, origin and their included organic remains. In regard to the origin he says: "The rivers which discharge into the Severn estuary, draining, as they do, a catchment basin of 9193 square miles, are the chief sources of supply" (264, 611). A source of secondary, but by no means slight importance is the sea, which has worn off material from the cliffs and which has carried muds into the Severn. As Sollas has fully explained, though the details cannot here be given, a small part of the silt which is brought down by the rivers may be deposited in the estuaries themselves, but the greater portion is carried seaward, "so that the final resting-place of the sediment of the Severn is situated some distance out to sea." A microscopic examination of the muds from a large number of localities on both sides of the Severn and along its tributaries revealed the following organic remains: "Coccoliths and rarely coccospheres, both of the ordinary cyatholith type so common in adjacent seas and in the Atlantic ooze; Foraminifera such as Miliola, Textularia, Nonionina crassula, Polystomella umbilicata, Rotalia sp., Spirillina sp. . . . spicules of Alcyonaria rarely; fragments of Echinoderm skeletons and minute spines: and triradiate spicules of Calcisponges, probably derived from Sycandra ciliata and S. compressa. The siliceous constituents are chiefly sponge-spicules, very rarely Radiolaria, and a variable quantity of Diatoms." The remarkable feature about these remains is that they are all marine, and yet they sometimes occur on the banks of the rivers at a great distance from truly marine waters. Moreover, the remains which are found are of organisms not living within many miles of the places where they occur, for Sollas has carried out a careful investigation of the fauna along the coast. He says: "Sponges do not grow anywhere so near Bristol on this side of the Channel as Portishead and Weston; Lynton, which is about 60 miles away, is the nearest possible locality; while Ilfracombe, about 15 miles further west, is well known as a rich collecting ground for both siliceous and calcareous sponges, and a host of other marine forms, including sea urchins and starfish, which might well furnish the echinoderm network and spines so frequent in the ooze. On the other side of the channel one would need to go to Bridgend before meeting with much in the way of shore life, and I doubt, after a hasty visit to that locality, whether much would be found there; a good deal farther west is Tenby, and no naturalist needs to be informed of the luxuriant growth of all kinds of marine animals, including sponges to be met with there" (264, 619). Sollas clearly shows that the remarkable distribution of marine remains far up the estuaries is due to the strong tidal current which brings the debis of organisms living along shore in the more open waters. This current rushes up the Severn at the rate of 6 to 12 miles an hour and it distributes the muds and microscopic organic remains as far north as Gloucester and to every estuary opening into the Severn. "On these shores, so remote from their source, some of these organic fragments find a permanent resting place, and thus far inland we discover along a river bank deposits, containing marine remains. But those which stay are few compared to those which are washed away again and carried out to sea, there to be deposited in marine mud-banks, probably not far from their original home" (264, 620). Just a few miles above the points at which the marine organic remains were found, the muds were examined and every sample showed abundant sponge-spicules, but these all proved to be of the fluviatile species, Spongilla fluviatilis, and none of the marine forms were found. This is the fauna of the recent muds, but a section through the older alluvial deposits which have a maximum thickness of 50 feet, shows that conditions have been much the same for a long time, and that there has been a constant alternation of conditions, first marine, then terrestrial, with the formation of peat beds. Wherever the estuarine sands and muds are washed over the peat beds a similar fauna, dominantly marine, though with fresh-water forms intermixed is found. The section is as follows in descending order:

Zone 1. Upper clay.	$\left\{{\begin{aligned}\\\\\end{aligned}}\right.$	a. More sandy zone, 5 to 7 feet
		b. More argillaceous zone, with disseminated vegetable matter 7 to 8 feet,
Upper peat, 1 to 2 feet, 6 inches.
Zone 2. Lower clay
Lower peat, 1 to 4 feet
Zone 3. Sands and mud
Gravel
Triassic sandstones

The deposits in the Severn estuary indicate a gradual subsidence of the land or advance of the sea. Both the upper and lower clay are blue and usually highly fossiliferous. In some sections a few feet will show an abundance of Foraminifera followed by several feet containing vegetal matter. In one section in which no peat was exposed the shaft sank through 39 feet of clay and then struck a marl bed one foot thick containing Limnæa, Planorbis, Scrobicularia piperata, Cardium edule, diatoms and Chara.

Yet another illustration of the nature of the faunas of estuaries may be found in the complete lists given in Verrill and Smith's invaluable report on the Invertebrate Animals of Vineyard Sound and Adjacent Waters (280). The fauna as there recorded from the sandy shores and bottoms of estuaries of the southern New England coast includes: Insects, Crustacea, Annelids, Gastropods, Pelecypods and Nemerteans all of which, with the exception of the last group, are represented by eight or more species (280, 170, 171); from the bottoms of sheltered estuaries, ponds, and harbors the following fauna is noted as characteristic:

Insects (4 species), Crustacea (30 sp.), Annelids (13 sp.), Nemerteans (2 sp.), Nematodes (2 sp.), Gastropods (15 sp.), Pelecypods (18 sp.) (Verrill and Smith 280, 176–178.) The study of the brackish water bodies in the region just mentioned has shown that the animal life is very abundant and that the number of species found, while not so great as in the open sea, is still fairly large. Particularly is it to be noted that the species which do occur are abundantly represented and are remarkable for their hardiness and ability to live under widely varying conditions. A few of the species are restricted to the brackish water, but by far the largest number are able to live in pure sea water.

SUMMARY OF FAUNAL CRITERIA FOR DETERMINING THE TYPE OF AN AQUEOUS HABITAT

The chief faunal characteristics of recent aquatic bionomic realms may now be summarized.

1. The typical marine fauna is widespread, large, with an abundant representation in individuals, species, and genera from practically all of the phyla of the invertebrate animal kingdom. The various lithologic facies have their peculiar faunules, but each one of these contains types from all or nearly all of the phyla, while many cosmopolitan species, particularly among the pelagic plankton and nekton are entirely uninfluenced by the substratum and their remains consequently will be found with those of forms restricted to particular facies. Furthermore, terrestrial and fiuviatile organisms will quite frequently be found in the marine fauna, having been transported there alive or dead, and their hard parts will be preserved with those of the typical marine species.

2. The number of genera and species in fresh water is as a rule very much smaller than that in the neighboring sea, although the number of individuals may be nearly as large. Entire classes of organisms are wanting, while other classes are represented by only a few genera and species. Fresh water organisms are distributed by rivers and are found living in lakes and lagoons on the subaërial portions of deltas, in lakes, playas and more rarely epicontinental seas, all of which water bodies are geologically short lived.

3. Brackish waters may be considered under two types:

(a) The brackish waters or land-locked epicontinental seas such as the Baltic show a range in salinity from that of normal sea water to that of the rivers which empty into such water bodies. The fauna is made up of a modified marine and a modified fresh-water fauna, and always has its nearest relatives in the open sea, on the one hand, and in the rivers and connecting fresh-water lakes, on the other. Since the species contributed by the marine waters are so much more numerous than those contributed by the rivers, the greater number of the species in the brackish water will be related to marine forms.

(b) The brackish waters of estuaries are not stable enough to have a fauna which may be considered endemic. The marine fauna lives along shore and in the waters not too much disturbed by the tidal current, but the organic remains found in the estuaries are marine. Moreover, along the coasts affected by the deposition of the tidal muds, no organisms live and it is only many miles away from the estuary proper that the marine forms are found whose hard parts, carried by the tidal currents and in time comminuted, finally come to settle on the floors of the estuaries and on the river banks. These hard parts are carried up the estuaries as far as the tidal current is felt and it is only above this point that the fresh water forms are found. In the very small area between the purely fresh and the dominantly marine waters is the brackish-water area in which there may be a small mixed fauna.

APPLICATION TO THE PAST

Armed now with a considerable wealth of facts drawn from the present, we may turn to the past in an attempt to set forth any available criteria which may be used in determining the nature of a given habitat.

Marine Deposits and Faunas. Sediments which accumulated in the open marine waters at all times, subsequent at least to the Pre-Cambric, have been found to contain a rich and varied fauna in which were represented all of the larger groups of the invertebrate animal kingdom which are recognized today. One need only mention the prolific faunas of the Cambric of St. John, New Brunswick, and of British Columbia, the Trenton of New York, the Niagaran of New York and elsewhere, the Hamilton of the eastern United States, the Muschelkalk and Upper Jura of Germany, the Upper Cretacic of the middle and north of Europe, and the Eocenic of France and England. Not only is the number of individual fossils great, represented by many species, but these species are scattered through many phyla, just as at the present time the organisms in the oceans are numerous and diversified, no one class reigning to the complete exclusion of others. This does not mean that we shall find the same distribution according to phyla in the past, but we do know that it will be diverse. The vertebrates, for instance, cannot be of importance in faunas until their evolution has had time to take place, and thus they are not found represented in the rocks in abundance before the Devonic. Thus the important phylum of Pisces find no, or only rare representation in the early Palaeozoic rocks; but, on the other hand, there were the Crustacea throughout the Palaeozoic, especially the trilobites, which became extinct at the end of that period. And so one might nicely appose the phyla, or more often orders or families, which were represented in the past, but are not now, and in this way we would see that the past, though different from, was similar to the present, and that Palaeozoic seas, even the earliest ones, lacked not in life and in the diversity thereof.

The very nature of marine waters, their continuity and great extent, suggests migration and wide distribution through currents. Barriers there were, of course, both by land masses and ocean currents, streams of cold water, and so forth, but, nevertheless, we know that migration along the coasts of the continents took place as it does today and that many species or at least genera spread throughout all of the oceans, for if we did not believe in the forces of migration and dispersal we would not have laid down the laws of correlation which are universally recognized. In no way, then, can a typical marine fauna remain bottled up in one place, with none of its members escaping to adjacent waters; such a thing cannot happen today and it is not reasonable to suppose that it happened in any geological period in the past.

Fluviatile Deposits and Faunas. The importance of sediments containing river faunas has not heretofore been realized, nor have such sediments and their characteristics been dwelt upon by most geologists. Grabau has been the staunchest advocate of the fluviatile origin of many deposits both in this country and Europe, but has usually stood alone in his interpretations. In his paper on "Early Palæozoic Delta Deposits of North America" he has described in great detail a large number of delta deposits occurring in the Ordovicic and Siluric and has shown what are the characteristics physical and faunal of such deposits. Barrell has likewise made a number of contributions to the study of fossil delta deposits with especial emphasis on their physical characteristics, and on the climatic factors controlling sedimentation.

It is a matter of difficulty to determine much about rivers of older geological periods, because the river channels are seldom preserved, especially in the Palæozoic, and when found are visible, usually only in section and cannot be traced along the surface. Flood plain and delta deposits are almost the only records of their presence left by ancient rivers. It is not to be expected, however, that such deposits will be without fossils any more than similar deposits today are. Rivers carry large amounts of detritus varying in grain from fine muds, a fraction of a millimeter in diameter, to bowlders often several feet across, though these coarser elements are, more likely to be carried by torrential or mountain streams than in the larger rivers. In a pluvial climate this load is brought into lakes or to the ocean and there deposited; in an arid climate it is spread out on interior plains or in basins in the form of alluvial fans or dry deltas. Since the lithological characteristics of deltas and flood plains are often of great assistance in the recognition of fossil deposits of this type, it may not be amiss to say a few words about them here.

The sediments spread out by a river in its lower reaches are of two types: (a) those which form directly at the mouth and are spread out in front of it into the sea, and (b) those which are spread out laterally either over the subaërial portion of the delta or along the flood-plain and over the neighboring lowlands throughout the lower portions of the river. These are the fine mud deposits of which we see such splendid examples in the case of the Nile and Mississippi deltas. The deposits in the Nile delta are thus described: "At low water these are visible in the steep banks which then rise 8 to 10 meters above water level. The hardened Nile mud forms a series of horizontal beds varying in thickness from a few inches to several feet, and looks more like an ancient stratified series than a modern deposit. The material of the Nile mud is a more or less uniformly fine-grained one, the size of the grains varying from ${\tfrac {1}{13}}$ to ${\tfrac {1}{100}}$ mm., rarely reaching to ${\tfrac {1}{10}}$ mm. in size" (Grabau, 87, 614). The Mississippi delta is spread out in the remarkable bird-foot form and the whole of its lower part is covered with a network of distributaries which often empty into large fresh-water lakes. In these lakes and over all the interstream areas the fine muds are deposited. They contain shells of fresh-water molluscs and much driftwood, which is often united into floating rafts.

In the portions of the delta nearer the sea, fresh-water and marine organisms are both found, not intermingled, however, but in separate layers, depending upon whether beds were deposited in the sea or by the streams above the sea. Thus a bed with fresh-water shells and lignite is often intercalated between beds with marine remains, giving evidence of the shifting conditions of deposition in deltas where streams continually change their channels and where consequently the areas of terrestrial deposition are shifted, while the sea advances in the interfluve areas and a wedge of marine deposits is formed.

Here a few details in regard to the nature of the Indo-Gangetic delta will give a good idea of what types of sediments and organic remains are to be expected. Lyell states that "No substance so coarse as gravel occurs in any part of the delta of the Ganges and Bramapootra, nor nearer the sea than 400 miles" (154, 280). A boring to a depth of 481 feet made near Calcutta showed below the surface soil, at the top of the first 120 feet of the boring, a stiff blue clay succeeded downwards by a sandy clay and this in turn by a peat bed. A nodular limestone, the kankar, of fresh-water origin, was encountered.^[3] Below the first 120 feet there were found various beds "consisting of clay, marl, and friable sandstone with kankar here and there intermixed, [while] no organic remains of a decidedly marine origin were met with . . . . The only fossils obtained in a recognizable state were of a fluviatile or terrestrial character. Thus, at the depth of 350 feet the bony shell of a tortoise, or Trionyx, a fresh-water genus, was found in sand, resembling the living species of Bengal . . . . At the depth of 380 feet, clay with fragments of lacustrine shells was incumbent on what appears clearly to have been another "dirt-bed," or stratum of decayed wood . . . . At a depth of about 400 feet below the surface, an abrupt change was observed in the character of the strata, which were composed in great part of sand, shingle, and boulders, the only fossils observed being the vertebrae of a crocodile, shell of a Trionyx, and fragments of wood very little altered, and similar to that buried in beds far above" (154, 281). This boring was very evidently through the subaërial portion of the delta, which was deposited at a time when the land stood higher and when, probably, hilly areas now removed by erosion or covered by deposits supplied coarser material near the seashore. The variability in the types of deposits is shown and it is seen that neither nodular limestones or conglomerates imply the presence of the sea for their formation. The sediments of the present delta are all fine-grained, the coarse deposits being found only at the foot of the mountains. Moreover, the fine sediments are carried far out to sea. "The sea, where the Ganges and Brahmapootra discharge their main stream at the flood season, only recovers its transparency at the distance of from 60 to 100 miles from the delta" (154, 279). In speaking of the Mississippi river Lyell says: "The prodigious quantity of wood annually drifted down by the Mississippi and its tributaries, is a subject of geological interest . . . . as illustrating the manner in which abundance of vegetable matter becomes, in the ordinary course of nature, imbedded in submarine and estuary deposits" (154, 268).

When the enormous transporting power of rivers is considered, when we think of the amount and variety of sediments together with terrestrial and fluviatile organic remains annually brought down to the sea by rivers there to be mixed with the marine sediments and the organisms living in the sea, we find it not so difficult to realize that the same phenomena happened in the past. The wonder would be if such intermingling had not taken place, and one must indeed be surprised to note how seldom it seems to have come to pass in the Palæozoic. Even admitting that land vegetation was mostly of a primitive, easily destructible, non-vascular nature in the Palæozoic, we still must marvel that so few fluviatile and terrestrial forms were carried out into marine deposits.

If the above characteristics are kept in mind it will not be difficult to formulate a certain number of criteria which may be used in recognizing a fossil delta or flood-plain deposit. That portion of the delta adjacent to the mouth of the river will be characterized by an alternation of marine and continental deposits, and these will be recognized in ancient deltas by a lithological and faunal interfingering. The silts brought down by the river will contain the remains of the river fauna, while the submarine deposits, be they sandstones, shales or limestones will contain a marine fauna. The two types of deposits as well as the two types of faunas, though they interfinger, will be of a distinct and recognizable character as a rule, and the nature of the faunas, in regard to numbers of individuals and species, will be recognized by the characteristics listed on page 77 above. The deposits on the subaërial portions of the delta and laterally in the flood-plain areas will consist of fine silts. Along the river banks the coarsest of the silts will be deposited and with them the heavier organic remains if there are any, such as the shells of molluscs. But with the finer silts periodically spread out at times of flood far to either side of the river, will be carried only the lightest materials, probably only plant remains and the exoskeletons of the various fluviatile crustaceous animals. Such organic remains may be carried out in great numbers, and if quickly buried will be excellently preserved in the fine muds. On the other hand, if they are carried a long distance, dropped and exposed to the air, and later perhaps picked up by some distributary and carried on again, the process being often repeated, they may be broken up, and when they finally come to rest and are buried, not a single complete organism will remain. Indeed, if in their final resting place they are exposed to the air for a long time and the mud on which they lie becomes sun-cracked, the fragments, drying up, may be blown for great distances, perhaps far inland or perhaps out to sea, coming to rest at last in regions far removed from those in which the organisms had lived and there amidst a strange fauna the remains may be entombed.^[4] In the sediments such a history could be read, if in shales or waterlimes the only organic remains were those of light specific gravity. Of the invertebrates there would probably be some arthropods or insects; among the plants, leaves, algæ, reeds and grasses would be expected. Such a deposit would be difficult to correlate with a marine deposit, because it might contain none of the contemporaneous marine organisms. If, perchance, some of the river organisms or fragments of them had been blown or carried to sea, their remains could be entombed with the typical marine fauna and the age would thus be determinable. It is not unlikely, too, that stray molluscan shells might be blown from the shore inland, as they are today, and might come to rest in the very muds in which the river organisms were buried. Likewise, in the low-lying portions of the flood plain near the sea, occasional high tides or inundations, through the wearing away of sand-bars, might allow the salt water to enter, carrying some marine shells into those regions. Then the fossil fauna would show a large number of forms belonging mainly to one phylum, the arthopods, and occasional single specimens of members from other phyla. If the opposite conditions prevailed, and the fragments of the arthropods or their exoskeletons were blown to sea, then the fossil fauna would reveal many marine organisms, complete and well preserved, from all or nearly all of the invertebrate phyla, and occasional fragments of another group of organisms which were not well preserved and whose occurrence in such surroundings seemed anomalous.

Brackish-Water and Estuarine Deposits and Faunas. It has been shown that at the present time there is no such thing as a brackish-water fauna made up of classes of organisms different from those found in neighboring marine and fresh waters. It might be extremely difficult to recognize from the sediments and fossils that any fauna had lived in brackish water, because unless the salinity had been reduced so much that it was nearly that of fresh water the fauna would not appear to be very different from a typical marine one, except that it would be dwarfed and would contain few species. An estuarine fauna would likewise be difficult to recognize from the fossils. These would, however, be likely to be fragmentary, even comminuted to microscopic size, and larger forms would be found only in the sands and coarser deposits along shore and not in the estuarine deposits proper. It has been seen that the conditions in an estuary are not favorable for supporting life. The tidal scour, the churning up of the water, keeping the sediments constantly in suspension, the sudden change in salinity twice every day, are environmental factors not at all conducive to attract marine animals which can find more stable and beneficial conditions along the coast on both sides of the estuary. Thus, we saw that in the Severn the organisms whose comminuted remains were found in the muds lived many miles away in the quieter waters north and south of the estuary. In the geologic column we shall probably rarely be able to recognize estuarine deposits from the faunas, but if at all it will be from the nature of the sediments, their lithological characters and sources.

THE EURYPTERID FAUNAS AND ASSOCIATED ORGANISMS

In the following lists the eurypterid faunas of the various occur- rences is given, as well as the organisms other than eurypterids which are found associated with them. No account is taken here of single occurrences or of the presence of a few fragments in normal marine faunas.

ORDOVICIC

Normanskill Fauna.
Eurypterids.
Dolichopterus breviceps
Eusarcus linguatus
Eurypterus chadwicki
Pterygotus ? nasutus
P. normanskillensis
Stylonurus modestus
Associated organisms. Seaweeds and graptolites.
Rhombodictyon
Climacograptus bicornis
C. bicornis var. peltifer
Cryptograptus tricornis
Dicellograptus gurleyi
Schenectady Fauna.
Eurypterids:
Dolichopterus frankfortensis
D. latifrons
Eurypterus ruedemanni
E. pristinus
E. ? stellatus
Eusarcus ? longiceps
E. triangulatus
Hughmilleria shawangunk
Pterygotus ? nasutus
P. prolificus
Stylonurus limbatus
Associated organisms.
In sandy shales, seaweed, Sphenothallus latifolium
In black shales, graptolites and trilobites
Climacograptus bicornis
Triarthrus becki

SILURIC

Earliest Lower Siluric (Ee₁.) Fauna of Bohemia.
Eurypterids.
Eurypterus acrocephalus
Pterygotus barrandei
P. beraunensis
P. bohemicus
P. nobilis
P. cf. problematicus
Graptolites.
Monograptus turriculatus
Other species

Upper Lower Siluric (Ee₂) Fauna of Bohemia.
Eurypterids. (All very fragmentary.)
Pterygotus barrandei
P. beraunensis
P. blahai
P. bohemicus
P. fissus
P. kopaninensis
P. nobilis
P. cf. problematicus
Slimonia cf. acuminata
Associated fauna.
The typical and abundant Upper Siluric of Bohemia.

Wenlock Fauna of Pentland Hills, Scotland.
Eurypterids.
Bembicosoma pomphicus
Drepanopterus bembicoides
Drepanopterus lobatus
Drepanopterus pentlandicus
Eurypterus conicus
Eurypterus cyclophthalmus
Eurypterus minor
Eurypterus scoticus
Eurypterus, 3 sp. und.
Slimonia dubia
Stylonurus elegans
Stylonurus macrophthalmus
S. ornatus
Scorpion—Palaeophonus loudonensis
Ceratiocarid—Dictyocaris ramsayi (Taxonomic position doubtful)
Sponge—Amphispongia sp.
In beds above or below eurypterid layers, the following fauna has been found:
Graptolites
Dictyonema venustum
D. (Chondrites) verisimile
Cyrtograptus murchisoni ?
Monograptus priodon
M. vomerinus
Coral—Favosites sp.
Asteroidea—Palasterina sp.
Crinoids—fragments
Brachiopods—
Lingula lewisi
L. symondsi
Strophomena walmstedti
Gastropods—Euomphalus rugosus
Cephalopods.
Orthoceras angulatum
Gomphoceras ellipticum
Conulariida—Tentaculites tenius
Conularia monile
C. sowerbyi
C. sp.
Problematic—Nidulites favus.

Shawangunk Fauna of Eastern North America.
Eurypterids.
Dolichopterus otisius
D. stylonuroides
Eurypterus maria
Eusarcus cicerops
Hughmilleria shawangunk
Pterygotus globiceps
Stylonurus cestrotus
S. myops
No associated organisms
Pittsford Fauna of New York.
Eurypterids.
Eurypterus pittsfordensis
Hughmilleria socialis
H. socialis var robusta
Pterygotus monroensis
Stylonurus (Ctenopterus) multispinosus
Crustacea.
Ceratiocaris praecedens
Emmelezoe decora
Pseudoniscus roosevelti
Fossils in dolomite partings but not in the black shales and not associated with the eurypterids.

	Genera
Graptolitida	1
Annelida (denticles)	3
Brachiopoda (Lingula)	1
Pelecypoda (Pterinea cf. emacerata)	1
Cephalopoda (Orthoceras and Gomphoceras)	2
Ostracoda (Leperditia scalaris)	1

Bertie Fauna.
a. Bertie fauna of Erie district.
Eurypterids.
Dolichopterus macrochirus
D. siluriceps
Eurypterus lacustris
E. lacustris var. pachychirus
E. pustulosus
Eusarcus scorpionis
Pterygotus buffaloensis
P. cobbi
P. grandis
Associated forms.
Cephalopods—Orthoceras undulatum
Trochoceras gebhardi
Brachipod—Lingula sp.
Ostracod—Leperditia alta
Pelecypod—Goniophora sp.
Pulmonate Gastropods—Hercynella buffaloensis
H. patelliformis
Graptolites—"Buthrotrepis lesquereuxi" (formerly considered a seaweed, now identified by Ruedemann as graptolites)
Ceratiocarid—Ceratiocaris acuminata
Plant—Chondrites graminiformis (may be a graptolite)
b. Bertie fauna of Herkimer district.
Eurypterids.
Dolichopterus macrochirus
D. testudineus
Eurypterus remipes
Pterygotus macrophthalmus
P. cobbi
Associated forms.—Scorpion
Proscorpius osborni

Kokomo Fauna.
Eurypterids.
Eurypterus (Onychopterus) kokomoensis
E. ranilarva
Eusarcus newlini
Stylonurus (Drepanopterus) longicaudatus
Associated forms.
Ceratiocarids

Upper Siluric Fauna of Oesel.
Eurypterids.
Eurypterus fischeri
E. fischeri var. rectangularis
E. laticeps
Pterygotus osiliensis
Eusarcus simonsoni
Ceratiocarid—Ceratiocaris
Fishes (Celphalaspid)
Thyestes verrucosus
Tremataspis schrenldi
Crustacea (Hemiaspidæ)
Bunodes lunula
B. rugosa
B. schrenkii
Synxiophosuran: Pseudoniscus aculeatus
Ostracod: Leperditia sp.
Cephalopod: Orthoceras tenue

Temeside Fauna of England.
While this fauna is sparingly represented in a number of beds in this group, all of the species occur together at only one horizon, namely, in the Olive shales below the Temeside Bone-Bed. Unless otherwise indicated, forms are abundant.
Eurypterids.
Eurypterus acuminatus (r)
E. pygmaeus
E. spp.
Pterygotus banksii
P. gigas
P. ludensis
P. problematicus
Parka decipiens (eggs)
Crustacea.
Beyrichia kloedeni
Leperditia phaseolus var. gracilenta (r)
L. small species
Physocaris vesica
Plantae.
Pachytheca sphaerica (cc)
Pisces.
Auchenaspis salteri (r)
Cephalaspis murchisoni (r)
Ctenacanthus (r)
Onchus murchisoni
O. teniustriatus
Brachiopods.
Lingula cornea (cc)

Ludlow Fauna of Scotland.
Localities in Lesmahagow inlier.
(1) Along the banks of Logan water in Ceratiocaris beds
Eurypterid: Slimonia acuminata
Ceratiocarids.
Ceratiocaris laxa
Ceratiocaris longa
Ceratiocaris papilio
Ceratiocaris stygius
Ceratiocaris cf. murchisoni
Worm tracks.
(2) In same bed ${\tfrac {1}{2}}$ mile distant:
Eurypterids:
Pterygotus bilobus
Slimonia acuminata
Ceratiocarids:
Ceratiocaris sp.
Dictyocaris ramsayi
Coelolepid fish: Thelodus scoticus
Myriopods ? impressions of
(3) From Long Burn, tributary of Logan Water in same bed
Eurypterid: Pterygotus bilobus
Ceratiocarid:
Dictyocaris ramsayi (Taxonomic position doubtful)
Ceratiocaris sp.
Ostracods: Beyrichia kloedeni
Beyrichia kloedeni var. torosa
Pelecypods: Modiolopsis nilssoni
Orthonota sp.
Brachiopod: Lingula minima
Gastropod: Platyschisma (Trochus) helicites
Worm tubes: Spirorbis sp.
(4) One half south of Logan House in " fish-band"
Eurypterid: Slimonia acuminata
Ceratiocarids: Ceratiocaris longa
C. murchisoni
C. papilio
C. stygius
Physocaris sp.
Coelolepid fish: Thelodus scoticus
T. planus.
Fish fragment undet
Myriopods: Archidesmus loganensis
(5) At Logan Water in Pterygotus beds overlying Ceratiocaris beds.
Eurypterids: Eurypterus lanceolatus
E. obesus
E. scorpioides
Pterygotus bilobus
P. bilobus var. acidens
P. bilobus var. inornatus
P. raniceps
Slimonia acuminata
Stylonurus logani
Ceratiocarid: Ceratiocaris papilio
Synxiphosuran: Neolimulus falcata
Note: In same locality Pterygotus and Slimonia are found in abundance associated with Beyrichia kloedeni and Ceratiocaris.

Lanarkian Fauna of Scotland.
Localities in Hagshaw Hills anticline
Glauconome layer in bed 9
Eurypterid: Eurypterus dolichoschelus
Coelolepid fishes: Lasanius problematicus
Ateleaspis tessellata
Bryozoan: Glauconome disticha
Sponge.
Worm tube: Spirorbis sp.

DEVONIC

Old Red Sandstone Fauna of Scotland.
a. Pterygotus beds of Carmylie.
Eurypterids:
Pterygotus anglicus (cc)
Parka decipiens (eggs)
b. Acanthodian beds of Turin Hill (Arbroath flags).
Eurypterids:
Eurypterus brewsteri
E. pygmaeus
Pterygotus anglicus
P. minor
Stylonurus scoticus
S. ensiformis
S. powriei
Fishes:
Mesacanthus mitchelli
Ischnacanthus gracilis
Climatius scutiger
C. uncinatus
C. reticulatus
C. macnicoli
C. grandis
C. gracilis
Parexus recurvus
P. falcatus
Euthacanthus mitchelli
E. elegans
E. curtus
Cephalaspis asper
C. lyelli
Thelodus pagei
Plants:
Pachytheca, etc.
c. Old red sandstone of Lorne
Eurypterids:
Pterygotus cf. anglicus
Fishes:
Cephalaspis lornensis
Mesacanthus
Thelodus (?)
Ostracods:
Aparchites
Isochilina
Beyrichia
Chilognathous myriopods:
Kampecaris
Archidesmus
Plants:
cf. Psilophyton
d. Upper old red sandstone fauna of Ireland
Eurypterids:
Eurypterus hibernicus
E. scouleri?
Fluviatile pelecypod:
Amnigenia (Anodanta) jukesii
Fishes:
Coccosteus
Plants:
Archaeopteris
Bothrodendron
Calamites
Sphenopteris
Stigmaria
Ulodendron

In the preceding lists only those faunas have been given which are abundantly represented in species and individuals, and which may, therefore, be considered characteristic. Of all the faunas cited, that from the waterlimes of Oesel, while not containing the largest number of species, is yet the one which is preëminently representative. The eurypterids undoubtedly lived and died in the muds now forming the waterlimes, the remains which are found there having suffered practically no transportation, as we may judge from the perfection of preservation and the entirety of individuals. The specimens of Eurypterus fischeri occur in greater numbers and in a more perfect state than do those of any other known species, and the exoskeletons are not only not compressed, but they even show the original chitin and specimens can be removed from the rock almost entire; the surface sculpture and internal structure are as clearly visible as in a Limulus buried in the sand but yesterday. We have here, if anywhere, a representation of the normal habitat of the Eurypterida, and likewise the normal faunal associates. The analysis of this fauna shows that besides the eurypterids there are a number of crustacea which are commonly found with the merostomes, but never in a typical marine fauna, two species of fish of the type characteristic of the Old Red Sandstone, an ostracod and Orthoceras tenue. The presence of this single cephalopod has been considered by some authors to be so important that they would brand the whole fauna as a modified marine one, because of it; and yet the startling and commonly neglected fact is that this thin shelled Orthoceras is most evidently out of place, for while the eurypterids are so marvellously preserved, this one rare cephalopod is worn, macerated, and flattened into a tenuous, carbonaceous film, and thus there is no doubt that it was transported from its normal habitat and came probably as a dead shell into the region where the eurypterids were living. Its presence is truly of great importance as being the very exception which proves the rule that the eurypterids were not normally marine.

A glance at the components of the fourteen faunas listed shows that there is not a single case in which several species of eurypterids are found in a fair state of preservation in such numbers as to be considered a recognizable faunule—there is not a case, to repeat, in which the faunule, including all of the organisms represented, can be considered either marine or modified marine, that is, brackish or estuarine. The most constant associates of the eurypterids from the earliest Siluric on are certain peculiar crustaceans, Ceratiocaris, and the like, which are never found with the molluscs, brachiopods, and trilobites which are characteristic of marine faunas. The oldest scorpion known comes from beds carrying eurypterids, similarly the earliest fluviatile pelecypod and the first myriopods were also found in eurypterid formations. In North America, England, Scotland, and on the continent the forerunners of the Old Red Sandstone fishes, now almost universally recognized to be fluviatile, are found in the Siluric with the eurypterids, crustacea and spores of land plants, but not in the beds carrying typical marine fossils.

In the Bertie waterlime, which is second only to the waterlime of Oesel in importance, a large eurypterid fauna is found with abundant Ceratiocaris, two species of pulmonate gastropods, a problematic plant, and a few very poorly-preserved, marine fossils, which last, by their very scarcity and by the evidences which they show of having been transported, argue more strongly for than against the extra-marine habitat of the well-preserved eurypterids and Ceratiocaris.

The application of the criteria for the recognition of the types of fossil faunas and habitats shows beyond any doubt that the eurypterids, so far as we now know, never lived in the sea or in any partially or wholly detached portion thereof; the only possible type of fauna to which the eurypterids could have belonged was that which dwelt in rivers, and this is nowhere more clearly shown than in the Siluric, which marked the acme in development and universality of distribution for the Eurypterida.

---

1. Some parts of the following classifications were presented by the writer at the 1914 meeting of the Paleontological Society of America.
2. The examples and salinities in this table are taken from the tables compiled from various sources by Professor Grabau, and given in the "Principles of Stratigraphy" (87). The classification is new.
3. For description of kankar, see Grabau's Principles of Stratigraphy, 87, pp. 586, 719.
4. In this connection it is interesting to record that the eurypterids of Oesel have such a thin test, that specimens exposed by the breaking of the rock are not uncommonly blown away by the wind.