NO. 1431. BREEDING HABITS AND EGG OF PIPEFISH— GUDGER. 461 



continued into and lining the poach as an epithelium with many 

 mucus-secreting- cells (sec Lilljeborg- on this subject). This epithelium 

 becomes folded to form ^* nld.r'' for the eggs, with the membranes of 

 which, since there is no zona radiata, it comes in veiy intimate con- 

 tact, proliferating to till all interstices Ijetween. Into these prolifera- 

 tions blood vessels, forming dense networks, penetrate and form a 

 virtual placenta by means of which the eggs and embr5^os are provided 

 with oxygen and food through osmosis. The lips of the pouch are 

 cemented by a gummy secretion, which at the same time keeps out 

 the water and enables them to withstand the pressure as the young, 

 surrounded b}' a clear serum-like fluid, grow and distend the pouch. 



In his efl'orts to determine the time of hatching and the age of the 

 embryos, Huot took fresh-laid eggs from the pouch and put them into 

 running- water. This he also did with embr3'os ranging from early 

 stages up to those with vitellus nearly gone and almost ready to hatch, 

 but in all cases they died within forty-eight hours at the utmost. He 

 also tried in vain to introduce eggs into the pouch. He concludes that 

 the eggs of S. dumerilil are fertilized at the time of transfer. His 

 work on the development is confined pmctically to organogen}" in the 

 late larva? and in the 3'oung. He confirms Couch, though ignorant of 

 his work, as to the thick and thin walled parts of the swim bladder. 

 The 3'oung fish when hatched has a " notable reserve vitellus inclosed 

 within the skin of the belly." 



Two years later (190-i), Ludwig Cohn, working on S. tijplile^ 

 reviewed Huot's work on the marsupium. In thin sections, through 

 the region of the marsupium, under the oil immersion lens, he finds 

 that these eggs have a zona radiata^ that the skin-epithelium is con- 

 tinued into the whole of the pouch and surrounds the eggs save wdiere 

 these are in contact, and that there are mucus-secreting cells in the 

 outer but none in the inner epithelium. He ascertains that only the 

 connective tissue of the pouch contains blood vessels, and that the peri- 

 vitelline space is filled with the albuminous fluid which Huot noted. 



Cohn finds that the lining epithelial cells have "i^j>/feew'"'-like pro- 

 cesses, and that these penetrate the pores of the zona radiata. Hence 

 he concludes that food stufl' and oxygen are transmitted to the perivi- 

 telline space by osmosis through these slender pseudopods, and that 

 in this way the 3"0ung are nourished. He notes that at the pole of the 

 6g'8"7 where the embiyo is formed, the epithelium is folded into glands 

 whose mouths abut onto the adjacent zona radiata. He finds, however, 

 that there is no definite position for the germinal disk. In S.jloridce.^ 

 eggs have been noted with the germinal disk turned downward — that 

 is, toward the folds of skin forming the pouch, and upward — that is, 

 toward the bod}- of the fish. 



The work of Cohn, confirms and extends that of Huot, and the two 

 together show that the older writers were correct in their vague ideas 

 Proc. N. M. vol. xxix— 05— 30 



