Morphological Variation and Its Causes in A. tigrinnui 29 



and figure 5, had it been taken at the proper angle, would have 

 shown more. In adults, figure 3 on plate III shows a head in 

 which this depression is very marked, while figures 4 and 5 show 

 contrasting heads, larval and adult, of an extreme opposite type. 



I will explain these latter in connection with the next feature, the 

 gape of the mouth. 



Perhaps nothing varies so much as does the gape of the mouth, 

 and with its variation is correlated, directly or indirectly, nearly 

 every other, variation in the form of the head. To discover the 

 causes of many of these features was for a long time utterly im- 

 possible. Such variations as those already described were ob- 

 viously due to the causes ascribed. One had but to observe the 

 rate of growth land the resulting forms, together with certain 

 general habits of the larva, and the connection became plain. 

 Season after season similar results followed from similar condi- 

 tions, and vice versa. Certain cases, however, were much more 

 puzzling, and even offered, in some instances, results that seemed 

 to contradict the principles above laid down. To illustrate by a 

 typical instance : larvae i and 3 on plate I and larva 2 on plate 



II are from the same season's catch in the same pond ; adult 2 on 

 plate III and adult 2 on plate IV show two of these same speci- 

 mens after metamorphosis. It will be observed that they had 

 very large heads, both broad and long ; the gape of their mouths 

 was no less striking. Perhaps more striking still, though not 

 shown on the plates, were adults resulting from the more slender 

 larvae, like figure 3 on plate I. The heads here were almost 

 equally large and obvio,usly out of proportion to the bodies. Now 

 during this same summer I raised from the egg a number of 

 specimens in the laboratory, feeding them, in very large battery 

 jars, on the same food — daphnids and chironomous larvae — which 

 the aforesaid larvae were eating in the pond. This food I netted 

 daily from the pond and supplied to them in unlimited quantities. 

 However, despite the same food, the laboratory specimens soon 

 diverged widely from either the bottom-era wders or the free-swim- 

 mers in the pond. They throve excellently, though, despite the 

 concentration of their food supply, they grew more slowly. This 

 was accounted for bv their constant movement ; thev were ex- 



225 



