HOW A NATURALIST IS TRAINED. 635 



be found congregated at the lightest side of the dish. A lamp seems 

 to exert the same fascination on them as on the moths of a summer's 

 night. The student, armed with a magnifying-glass, now picks out 

 from the dish the forms he desires to study by means of a medicine- 

 dropper, and transfers them to a separate dish, where they may have 

 an abundance of water, and, when sufficient material has been picked 

 out, the real study begins. 



The first thing is to ascertain everything of the external and inter- 

 nal structure that can be seen in the living animal. For this purpose 

 it is placed on a glass slide in a drop of sea-water and carefully studied 

 under the microscope. In this, as in all embryological work, drawing 

 is absolutely necessary. Pages of description will not take the place 

 of pictorial representation. After the whole is studied, then comes a 

 study of the different parts, drawings and notes being made of each. 

 An embryo is continually growing, and it becomes necessary to take 

 into account every stage of growth. The embryo of to-morrow will 

 be different from that of to-day, and the changes must be recorded. 

 Some of the embryos are therefore kept, the water being changed, as 

 often as necessary, and these serve for to-morrow's study, the draw- 

 ings of to-day furnishing a basis of comparison. In many cases it is 

 a comparatively easy task to rear embryos until the adult condition is 

 recognizable, but at other times it is found impossible to keep them in 

 confinement for more than two or three days. In the first case it is an 

 easy task to identify the forms studied, but in the other the difficulty 

 is considerable. Subsequent skimmings must be made in the hopes of 

 securing the later stages of development, while an endeavor to find the 

 animal which produces the eggs frequently meets with success. Com- 

 parison with the studies of other investigators is also an important aid 

 to identification. 



If, however, the eggs are taken directly from the parents, this 

 trouble is wholly avoided, although other difficulties are introduced. 

 Suppose, for instance, that one wishes to study the development of one 

 of the fishes, the first step is to obtain males and females with the 

 generative products ripe. A gentle stroking will serve to expel both 

 eggs and milt, and then these are mixed together and " artificial im- 

 pregnation " is affected. In the case of worms, oysters, and clams, the 

 same result can be obtained by mincing the generative organs of ripe 

 males and females, mixing them together, and then straining off the 

 larger portions which, by their decay, would soon pollute the water. 

 In the case of crabs and shrimps the eggs are borne attached to the 

 abdominal legs of the mothers, and by capturing these females an 

 abundant supply of material can be obtained. The parents can readily 

 be preserved alive in lobster-cars or similar contrivances, and furnish 

 eggs as they are needed. 



Artificial impregnation is a very valuable process, for, by its aid, 

 every stage in development may be obtained. Eggs and milt may be 



