NO. 1431. BREEDING HABITS AND EGG OF PIPEFISH— GUDGER. 455 



describe as cutaneous and outside the dermal exoskeleton. They are 

 the first who describe the slit-liivc opening- at the anterior end of the 

 niarsupiuni. The,v think that Rathlve mistook Scyj)hius^ which never 

 forms a pouch, for S. acns witii this sac in the very first stag-e of 

 development, and that, by imagination, he supplied the other stag-es 

 necessary to complete the formation. They do not see how anyone 

 could possibly have mistaken for a male a female with yellow eggs in 

 the ovary. 



William Andrews, writing in 1S60, says of S. typlde that the ova 

 liberated by the female are received into the abdominal pouch of the 

 male, who has power of expanding its flaps and of fastening the ova 

 b}^ a highly viscous secretion. He is the first to observe that the full 

 development of the ova forces open the pouch and liberates the young. 

 He finds 8. pequorem individuals clinging side by side to bits of Zox- 

 tera by their tails, in which position he thinks that the male is enabled 

 to attach the eggs to his abdomen. He says that X typlde and acux 

 swim with their tails, which fact is also noted by Weinland and others. 



S. acu^^ according to Jonathan Couch (1867), has developing ova in 

 the pouch from April to October, and is veiy retentive of life. 8. 

 foridce. is ver}" amphibian-like in this latter respect, swimming about 

 and even jumping out of the aquarium some time after its head 

 has been cut off. Couch anticipates Huot in discovering that the air 

 bladder has an anterior thick-walled and posterior thin-walled part. 

 He describes three adidt specimens of S. sequoveus. (?) with well- 

 developed dorsal and ventral fin-folds. 



In the same year (1867), Lockwood was so fortunate as to see the 

 delivery of young in sea-horses kept in acpiaria. One male stood 

 vertically in the water, and pressing the point of his tail against the 

 bottom of the pouch, forced the young out at its mouth. The other, 

 catching its tail under the edge of a winkle shell, pulled the body 

 downward, rubl)ed the pouch against the shell, and thus expelled the 

 young. This was repeated, with intervals of rest (the fish seemed to 

 tire easil}'), for six hours. In August, 190^, I had opportunity to see 

 the delivery of the young from the pouch of a niale lUppociunpus 

 hudsonius at Beaufort, but beyond a mere relaxing of the sphincter 

 muscle at the mouth of the sac nothing was remarked. 



Lockwood says that at the time the ova are received into the pouch 

 its walls are thick and well lined with fat, but that, when the young- 

 are excluded, the walls are only one-sixth as thick. Hence he con- 

 cludes that this fat serves as food for the young. He adds that the 

 walls again become thick, so that he was several times led to think 

 the pouch gravid when it was not. The writer was similarly deceived 

 once, even so far as to try to open the pouch, whose walls must have 

 been five or six times as thick as those of a breeding pipefish. 



To Lafont is due the credit for discovering the mode of transfer of 



