plants most important both for a knowledge of the part plants play in the work of the world, and in their relations to man directly.

5. Both structure and physiology should be studied, but the use of the compound microscope should be reduced to a minimum.

In understanding the part plants play in the world, and for the control of plant production a knowledge of physiology is more illuminating than a knowledge of structure. Structure is more easily used in laboratory work in training in exact observation, but physiological experiment offers most excellent opportunity to draw logical conclusions from observed facts. From the point of view of scientific methods, the experiment is more valuable than mere observation, since it leads to the formation of judgments from a set of facts brought to bear on the solution of a given problem, or on the answer to a given question. The physiology and the structure of a given organ should be studied together.

6. The main structural features of leaf, stem, root, flower, seed and fruit should be studied, with the chief differences between monocotyledonous and dictoyledonous plants. Some observation should be made on the finer details of the cell, protoplasm, cell wall, nucleus, chlorophyll bodies; the cells concerned in feltilization, the pollen grain and the pollen tube, with their contents and the embryo sac; some work should be done on the morphology of the bacteria, yeasts and fungi.

7. On the physiological side a knowledge of the water relations of plants, together with a knowledge of the nutrition of plants as related to soil fertility, probably gives the most control over crop production. How plants absorb, transport, and lose water, the amount necessary, available water in the soil and its conservation, are the leading ideas concerning the water relation. A knowledge of photosynthesis shows the place the green plants occupy in nature, as the energy storers and food manufacturers. A knowledge of foods in general, their kinds, their sources and their uses, illuminates one of the most important relations of plants to human life, and shows one of the fundamental similarities of all living things.

Growth and the condition necessary for it, especially the temperature and moisture relations, the oxygen requirements and the food supply should be studied. A study of the raw materials out of which the food is made may lead to an understanding of soil fertility and the requirements for soil fertilizers. The last should be demonstrated by some experiments.

Some study should be made of the responses to stimuli by which plants. adjust themselves to their environment, especially the responses to light and gravitation.

8. The physiology of bacteria, yeasts and molds should be demonstrated by experiments, and these should be supplemented by reading a good text book. I would suggest the following subjects for experiment: The distribution of bacteria and fungi in air, water, dust, earth, and food (milk, bread, etc.). Their effect in putrefaction, and in fermentation, as the spoil

ing of cooked peas, the souring of milk, the alcoholic and acetic acid fermentations, the making of bread;

Their necessity for organic food, as shown by the amount of growth with and without organic matter, contrasted with the capacity of green plants to make their organic food;

Their temperature relations, minimum temperature for growth, and its relation to preservation of food in cold storage, the optimum temperature for growth and its relation to the rapidity with which food spoils in summer, the thermal death point, and its relation to sterilization and pasteurization of milk, the preservation of other foods, or to making suspected drinking water safe; the fixation of nitrogen in root tubercles on leguminous plants, showing the difference when grown in soil with and without the nitrogen fixing bacteria;

The study of a few plant diseases produced by fungi or bacteria, as the rust of wheat, of the hollyhock, or of the carnation, the mildew of some common plant, the brown rot of plums or peaches, the scab of apples, some leaf spot diseases, the smut of oats or corn, the fire blight of the pear, the crown gall or knot of fruit and other trees, or whatever material is readily available.

Some of this last material might be kept as a permanent stock by mounting it under celluloid. Few of the text books for high schools give experiments with the bacteria. Bigelow's "Applied Biology" gives some good ones for molds and yeasts.

9. Every course in botany should include some acquaintance with plants out of doors. Either in the woods and fields, or along the streets and in the parks, or even in vacant city lots and about homes. The knowledge that is most valuable here is that which enables one to recognize species of trees, shrubs, or flowers as acquaintances or friends, rather than to have a definite number of pressed specimens all labeled with complete classification, or full technical descriptions in note books. Students who feel an interest in the latter sort of work may be encouraged to do it for themselves, but it is not the most valuable knowledge to be required of all. This work should be carried out largely for its recreative value to the average individual. It might profitably include a knowledge of weeds, of common cultivated plants, of trees and shrubs commonly used for ornament, as well as a knowledge of plants growing in woods and fields. It should include the recognition of those characteristics which make the plants undesirable, or, desirable for the uses to which they are put. Excursions, though only around the block to get acquainted with the trees planted along the street, may be made the basis of teaching, of testing and of drill exercises. The sort of knowledge desired comes only with frequently repeated observations of the same species.

In the relations of plants to their environment, both organic and inorganic, various ecological observations may be made. The struggle for ex

istence among weeds in a vacant lot or of weeds and crops in a cultivated field, may offer convenient material. In many ecological problems the work must as yet be largely observation, with little in the way of conclusion or general law. Methods of seed distribution, of pollination, observation of plant associations, and of plant successions, are a few of the ecological problems that may be studied in the field if the situation is favorable.

PHYSIOGRAPHY CONFERENCE

FIELD METHODS OF GLACIAL GEOLOGY.

FRANK LEVERETT, GEOLOGIST, U. S. GEOLOGICAL SURVEY, ANN ARBOR.

In the early and middle part of the 19th century the attention of field geologists interested in glacial features was directed largely to the question of the extent of the drift and evidence as to its mode of deposition, whether by floating ice or by land ice. The presence of buried soils was noted at an early date but was not recognized at first as evidence of repeated glaciation; instead it was cited as an evidence against land ice and in favor of iceberg deposition, for it was argued that land ice should have torn up all the old soil it overrode. The loess deposit which overlies a considerable part of the drift was early cited in favor of extensive submergence, though now it has come to be generally regarded as a wind deposit.

Not until the latter part of the century, when Chamberlin, Gilbert, McGee, N. H. Winchell, Upham, and others began systematic work on the drift, were moraines and other land ice features clearly recognized and the iceberg hypothesis effectually displaced. The evidence for successive glacial stages was also brought into clear recognition. Following this and largely through investigations by the Canadian Geological Survey, it became known that the ice had more than one center of dispersion. This was known for the last stage of glaciation and inferred for earlier stages. The location of ice in the great valleys, early brought to notice by Chamberlin, was soon seen to give rise to a complex lake history, to which Upham, Gilbert, Taylor, Spencer, and others had contributed important chapters by the end of the century.

It is fortunate that the glacial investigations by the United States Geological Survey were from the beginning directed by one with so broad a grasp of this intricate subject as is possessed by T. C. Chamberlin. Having made a general reconnaissance from the Rocky Mountains to the Atlantic seaboard prior to his connection with the Federal Survey, he at once outlined plans for a comprehensive mapping and study of the leading features

of the glacial deposits of the entire field. By this method the knowledge of our glacial deposits has grown symmetrically and there has been a rapid unfolding of the leading events or episodes of glacial history. In this work several of the State geologists have been in close touch with Chamberlin and have consulted with him as to methods and plans. Work done under the auspices of universities and colleges has also been brought into the same general plan. As a result there is already worked out a somewhat complete outline of glacial stages and an extensive mapping of moraines and other features for the large glacial area of northeastern United States. The subject is now advanced to such a stage that detailed areal study and mapping can be taken up and the results interpreted in relation to the broad questions and problems as well as to local ones.

In the development of North American glaciology under Chamberlin a number of cognate questions of economic value have been given atention; such for example as the relation of soil distribution to glacial history; the distribution of plant societies in relation to various types of glacial deposits and their proper successors in cultivated crops; the occurrence of underground waters and availability of such waters for public and domestic use; also locations and surroundings of valuable water powers due to glaciation. So also have questions of more purely scientific nature, such as changes of drainage and peculiarities of drainage development due to glaciation; recent uplift of parts of the glaciated region as shown by the inclination or warping of the shore lines of the glacial lakes, and the bearing of these on questions of the effect of ice weighting, and ice attraction; the rate of recession of water falls and excavation of valleys in their bearing upon the length of post-glacial time, and numerous other scientific matters of more or less consequence.

As outlined by Chamberlain the glacial investigations embrace a careful mapping of all moraines, outwash plains, lines of glacial drainage, eskers, kames, drumlins, and intermorainic till tracts, and all other features necessary to a complete general exhibit on a map of the structure and topography of the drift. It also embraces a study of natural and artificial exposures, of well records, and all available material bearing upon the succession of glacial formations.

In the areas covered by glacial lake waters they embrace a study of the several shore lines and their relations to lake outlets and also their relation to the ice sheet and its oscillating border; the lacustrine deposits over the entire lake areas are also studied sufficiently to show the extent of each of the leading types of soil. Inasmuch as earth warping or differential uplift has affected the shore lines of a considerable part of the lake areas the amount of tilting of each beach is more or less carefully studied in order to fix upon the time when uplift began, when it was most rapid, and in what directions the tilt lines trend. This study alone embraces hundreds or even thousands of miles of travel and a large amount of careful leveling to prop

erly interpret the nature of the uplift, its geologic relations, and time relations.

One of the first essentials to correct interpretations in glacial studies is the preparation of a complete map on the ground. It is not sufficient to note the width of a certain kind of deposit, where it is crossed by the lines traversed in the field, and then make up a map in the office, although such has been the practice of a number of glacial students. Such a map made up in the office is certain to be far less accurate than one on which the extent of each deposit is carefully noted in the field.

In the field one is often forced to draw upon every available line of evidence to support his mapping, because of the incompleteness of the exposures. Were the glacial geologist to publish merely an exhibit of what is shown in outcrop, as is frequently done by the stratigrapher, the map would be wholly unintelligiblee to persons not familiar with the particular field, even if somewhat well trained in glacial investigations. It is very necessary to determine the general drift structure in order to make an intelligible map. To do this in the absence of exposures the glacialist watches for every indication of a change of drift structure, in changes of the vegetation or character of crops, as well as in the materials brought up by burrowing animals. Constant inquiry of residents as to the soil and its relations to common crops must be kept up, and knowledge obtained of borings or excavations of all classes that have been made along or near the line of route. There is thus no opportunity for the glacialist to become absorbed in meditation or otherwise preoccupied unless he stops on his trail. He must too, with each change in drift structure, be alert in mind and ready to furnish an interpretation of the cause of change. Otherwise, the map will become a piece of uninteresting routine work that may soon become tedious and even fatiguing.

The mapping of boundaries between different classes of glacial topography and between different kinds of drift with similar topography is done by methods adapted to each feature or deposit. In tracing the course and the limits of such a strong topographic feature as a moraine, it is hardly necessary to leave the wagonroad, the mapping being easily done on topographic maps; while in the absence of such maps distance from section lines, streams, or other ordinary map features may be readily estimated. If a moraine is but a mile or so in width both borders may be traced in a single traverse by a zigzag course along it; but if several miles in width each border and also the crest may require separate tracing; the moraine is also usually crossed at intervals sufficiently frequent to familiarize one with the variations in expression.

The beach lines of the glacial lakes and to some extent the eskers must be followed at close range to insure correct mapping. Topographic maps are especially useful in giving these linear features their correct position. The same is true of drumlins and kames.