COLORATION" OF FISHES LONGLEY. 481 



Knowledge of this fact, and of the fact that at night these species 

 feed upon food of definite sort, whose range is known, permits one at 

 last to speak with some confidence regarding their distribution at 

 different hours. 



Important, however, as is the information gained through deter- 

 mination of the feeding habits of fishes, or of other animals, through 

 analysis of their stomach contents, the vividness of one's mental 

 picture of their daily round of activity is increased immeasurably 

 by direct observation of their behavior. Nothing could be clearer 

 than the import of masses of statistics showing that some fishes feed 

 by night and others by day. Yet his conception of the fact is pale 

 and colorless who has not seen the nocturnal species coming in at 

 daylight singly, or in twos or threes, until their accustomed school- 

 ing places swarm with them; or who has not noted, at dusk, forms 

 which have been active all day disappearing, in what way it is 

 almost impossible to discover, while others from the inner, dimly 

 lighted, secluded fastnesses of the reef, come out of obscurity some- 

 what before, but with as little confusion, and with almost as definite 

 an order of appearance, as the stars in the twilight. 



At many points the method of direct observation yields results, 

 however, which could never be inferred, even from complete knowl- 

 edge of the food of fishes and the conditions under which it is taken. 

 This is the case with all the minor diurnal activities of nocturnal 

 species; and is quite as true of the common changes of color, 

 shade, and pattern, which species of the most diverse sort effect 

 under a great variety of conditions. With respect to all such matters 

 it is indispensable, and the results to be obtained by pursuing it are 

 crucial in determining the significance of the coloration of fishes. 



To secure a comprehensive pictorial record of specific differences 

 in behavior it is practically necessary to use a camera inclosed in a 

 water-tight container more or less after the model of that shown in 

 figure 1, plate 2. The submarine photographs illustrating this article 

 were obtained with a 4 by 5 autograflex camera, protected as appears in 

 the picture. The camera " looks " through a circular window in the 

 left end of the box. What is in focus in the field of the lens appears 

 upside down in the mirror mounted in the upper end of the focussing 

 hood. The act of focussing, the tripping of the shutter, and other 

 operations necessary in making either instantaneous or time exposures 

 are made by various screws and plungers, some of which are not visi- 

 ble in the figure. It is necessary to send the box to the surface for the 

 purpose of changing the plate after each exposure. 



The difficulty with which pictures may be secured varies with cir- 

 cumstances. When the light is good an exposure of a tenth of a second 

 is sufficient; that is, a snapshot may be taken; and if the subject 

 is reasonably quiet and the water calm, it is not particularly 



