The experiments at Devil's Lake have shown clearly that many species of fish can withstand changes in density of about .011, that the ability of the fish to withstand such changes is largely dependent on the temperature and on the condition of the fish at the time, and that marked differences exist between individual fish in this respect.

The reasons for the disappearance of the pickerel can not be stated with certainty. Coincident with their disappearance a marked decrease in lake level took place. This cut off the coulée coming in from the lakes northward in which the pickerel bred. Previous to their disappearance they were taken from the lake in large numbers without restriction, and this fact coupled with the loss of their breeding grounds is sufficient in some measure to explain their disappearance. It does not, however, explain the suddenness of their disappearance, which, according to all reports, took place between 1885 and 1889. It is probable, therefore, that some epidemic was the third factor in their reduction. Regarding this it is of course impossible to obtain definite information now.

A second line of work conducted by the station is the study of the variation produced in organisms by changes in the physico-chemical character of their environment and by isolation, using for this purpose mainly the copepod Diaptomus (Fig. 8).

FIG. 8. Diaptomus (MALE) SHOWING PROCESS, α, OF THE RIGHT ANTENNA. The length of this process, together with that of the antenna and the body, shows marked differences in inhabitants of different ponds, as the result of differences in environment and of isolation. Magnified 50 diameters. (Moberg, del.)

A large number of these from ten of the separate parts of the lake collected in all seasons of the year have been carefully measured and compared. The following data have been chosen for such comparison: length of body, antennæ and process of right antenna in the male.

This study has shown clearly a distinct variation in all of these dimensions as the result of such changes in environment, the length of the process of the antenna of the male varying directly, while the length of the antenna varies inversely as the length of body.

These results are consistent with those of many other

workers respecting the influence of environment and isolation in modifying organisms. They open up a wide field of further observation and experiment, such as the transfer of organisms from one pond to another to determine whether the differences are constant or reversible; the gradual increase in density of the water to determine how soon such increase shows its effect, and whether it has any other effects in influencing the sex and number of offspring in successive generations, etc.; a comparison of the differences found between organisms in lakes far distant and those in the Devil's Lake complex; a study of the developing organisms in the various ponds to determine whether the larvæ show differences comparable with those of the adults; these are some of the many interesting and important problems waiting to be solved. Such questions, while seemingly of interest only to the specialist, are nevertheless of profound importance to all of us, for they seek to answer in some measure the ever-vexing question of "nature vs. nurture." If it is reasonable to believe that all protoplasm is fundamentally alike, however widely it may differ in its manifold forms and reactions, then the solution of this question of the relative part played by heredity and environment in the development of even so humble a creature as a little shrimp-like animal about one twentieth of an inch in length, should aid in answering the same question in human development. The fact that in the Devil's Lake complex we have in many cases an exact, and in all a fairly accurate record of the time that nature has been performing these isolation experiments for us, and that we have ready made, as it were, a large number of experimental ponds in which to conduct further experiments of our own, renders this region an unusually favorable one in which to study the problem.

The plankton, the minute free-swimming organisms, in the lake has been extensively studied in its relation to its environment. These studies comprise over 800 collections from all parts of the main body of the lake and several of the outlying ponds. These have been taken both by day and by night, at all times of year, including midwinter. The organisms contained in these collections are now being counted, after which a comprehensive report will be prepared. Coincident with many of these collections determinations of temperature, dissolved gases, etc., have been made. Studies of the plankton have brought out clearly the interesting fact that these minute organisms frequently occur in swarms, the causes of which are not at present clear. In two regions separated by only a few feet may

occur immense numbers, especially of the cladoceran Moina, while the intervening areas may be comparatively free from it. Differences as great as 400 per cent. may occur between two places seemingly alike in every respect. So far as can be determined, there is no difference either in chemistry, physics or food supply to determine these differences. Their explanation must therefore be sought in some as yet obscure biological factor.

Some preliminary studies have also been made of the Red and Missouri rivers and of other waters in North Dakota, but the main activities of the station have been confined to Devil's Lake and its adjacent waters.

The fauna and flora of the lake are interesting both in respect to what they do, and do not contain. The only vertebrate inhabitants are the stickleback, the leopard frog and the salamander Amblystoma. Among the invertebrates the crustaceans, insects and rotifers are each represented by several species. One of the latter Brachionus, named from its habitat satanicus by Mr. Rousselet of London, is new to science. A few extinct mollusks occur and at least one arachnid, but the cœlenterates, sponges, polyzoans and annelids are apparently entirely absent, and thus far but one free-living flatworm has been found. Nematodes on the contrary are common, and include several species as yet only partially determined. The animal phylum most largely represented is the protozoa. These include chiefly ciliates and flagellates, many of which are as yet indeterminate and probably represent new species. The rhizopods are poorly represented, Arcella vulgaris being the only one which occurs at all commonly. This is the more remarkable in view of the fine layer of ooze covering the lake floor. The fact that this layer is occasionally lacking in free oxygen may explain their absence, but Juday's observations on animals living in water lacking free oxygen, and the presence of several other protozoa in the ooze in Devil's Lake, render this explanation somewhat questionable. One of the commonest animals in the lake, the midge Chironomus, occurs as a larva exclusively in this ooze.

The ditch grass (Ruppia) grows abundantly in the shallower parts of the lake, forming a tangled mass, which in some places prevents navigation with power boats. There are many species of unicellular algæ; some of which, especially the bluegreens, are very abundant. It is difficult to assign to any one

4 Juday, C., "Some Aquatic Animals that Live under Anærobic Conditions," Trans. Wis. Acad. Sci., XVI., p. 11.

group of plants the most important part in the life of the lake, although the diatoms with some fifty species take first place in respect to variety and play a very important part as food for the animals.

The busiest part of the lake, biologically speaking, is the Ruppia zone which occurs in depths of from two to five feet. Here is the nesting place of the stickleback. Attached to the Ruppia are masses of the filamentous green alga Cladophora, while to this again are attached numerous sessile diatoms and protozoans. Many free-swimming protozoans, rotifers, nemaodes and copepods also occur here; some by chance as they are driven hither and yon by the currents in the lake, others finding favorable shelter in the tangled mass of Ruppia and Cladophora, and seldom occurring elsewhere.

The rocky islands are the breeding homes of cormorants, ducks and terns, while many other species of water fowl breed abundantly in the marshes, which in some places occupy the old lake bottom. The cormorant colony on "Bird Island," as it is known, is one of the most interesting places about the lake

FIG. 9.

BIRD ISLAND IN DEVIL'S LAKE, the home of the double-crested cormorant, whose nests may be seen among the rocks. (Young, photo.)

(Fig. 9). On the approach of the intruder the old birds, and such of the young as are able, scramble off into the water, while the birds which are too young to leave the nest voice their

protests in inarticulate squawking, at the same time disgorging masses of sticklebacks and young salamanders, which form. their staple diet. The nests, rude piles of sticks, are crowded together among the rocks. After the breeding season gulls, ducks and terns gather in flocks about the lake, and together with migrant phalarope and sandpipers form a busy and restless population. Some of the flocks of ducks are so large that as they take wing they make a sound like rushing water.

On the southern shore is a long low ridge of morainic hills, the highest point of which is named "Sully's Hill" after General Sully, the old Indian fighter (Fig. 10). The major part of

FIG. 10. DEVIL'S LAKE, shore strewn with glacial boulders in fore-ground. Sully's Hill National Park site in back-ground. (Young, photo.)

this hill which is some three miles long has been set aside as a national park and fenced in for the retention of wild animals (elk, deer, bear, etc.) with which it is planned to stock it.

About two miles to the west of Sully's Hill is the old frontier Fort Totten, now used as an Indian school, while in wretched "shacks" scattered over the hills on the reservation still live some of the old warriors, who in days gone by hunted the buffalo and the white man over the vast plains of the Dakotas.