226 SEXUAL DIMORPHISM 



Sub-order Lophobranchii. The chief secondary differ- 

 ence between the sexes in this Order is the pouch which 

 is commonly present in the males for holding the eggs 

 during their development. The pouch, however, is not 

 present in all the genera, nor when present has it in all 

 cases the same structure. In Nerophis, the snake pipe-fishes, 

 in which the body is smooth and rounded, the eggs are attached 

 to the skin of the abdomen of the male in front of the anus, 

 and not covered by any cutaneous folds, but exposed. In 

 Syngnathus and in Siphonostoma the eggs are contained in 

 a pouch situated beneath the tail behind the anus, and 

 formed of two longitudinal flaps which meet in the middle 

 line but are not united at their edges. In Hippocampus the 

 pouch is formed of similar folds united together by their 

 edges, an opening being left only in front. It is reasonable 

 enough to suppose that at the beginning of the evolution of 

 these forms the eggs were adhesive and became attached to 

 the ventral surface of the tail, except in Nerophis, where the 

 tail being very slender and prehensile the eggs were received 

 on the abdomen. In Syngnathus and Siphonostoma the tail 

 is square in section and its ventral surface is flat. It is 

 probable that the fish would exert itself by muscular effort 

 to form the lower surface of its tail into a groove in order to 

 retain the eggs. Such an effort might well be excited by 

 the tactile stimulation due to the presence of the eggs, and 

 would probably lead to the depression of the central part 

 and the increased growth of the skin at the edges of the 

 ventral surface of the tail. Granting that growth took place 

 in this case, as in other cases, in the directions in which 

 straining of the skin and muscles was habitually set up, the 

 evolution of the longitudinal folds is, on Lamarckian principles, 

 explained. The fusion of the edges of the folds in Hippo- 

 campus may be attributed to the edges being pressed together 

 with greater force. We have no reason to believe that 



,{ 



