466 PROCEEDINGS OF THE NATIONAL MUSEVM. vol. xxix. 



minal vesicle can not be seen. The sections 1 have made of eg^s just 

 (>xtviided are so unsatisfactorv and so little understood that further 

 investigation is necessary before sections are figured. The older 

 observers, Retzius (1833), Rathke (lS3f3, 1837, 1840), Vogt and Pap- 

 penheim (1859), although they studied the ovary with the microscope, 

 missed these peculiar structures. Later observers — Brook, McLeod, 

 Cunningham (1897), and Huot (1902) have made sections but have not 

 gone very far into the structure, nor will I myself do so now, since it 

 is ni}^ intention to work up the organization and development of this 

 organ later, the material for this being now on hand. 



II. THE METHOD OF DEPOSITION. 



This has already been described in the first part of this paper, but 

 it may be well to emphasize the fact that the process is such as to 

 prevent absolutely any contact of the eggs and sperms with the sea 

 water. 



III. FERTILIZATION. 



The &gg of S/pltostoiiia fiorklx,^ as before mentioned, possesses a 

 very thin and perfectly transparent shell. This surrounds an egg- 

 made up of straw-colored 3'olk having man}" orange-red oil globules 

 imbedded in its periphery and these surrounded in turn by a thin pel- 

 licle of protoplasm. The colored oil globules render the agg so 

 opaque that I have never ])een able to find the microp3de. Yet, 

 strange to say, the ^gg of a related European form, Syngnathiis 

 ophujlon^ was the first fish and possibly the first vertebrate Q,gg in 

 which this opening was discovered. Whether this q,^^ is transparent 

 or not I can not sa}^, but in it Doyere (1849) foimd the micropyle just 

 over the '''' dmiue jproligere^'''' and gave its diameter as ^\-^ mm. 



A. Natural fertilization. — Different investigators var}^ in their 

 conclusions, or, more correctly, their conjectures, as to the time of 

 fertilization. A jrriorl., one Avould expect the fertilization to be 

 efiected at the time of transfer. Probably the surest wa}^ to deter- 

 mine the time of impregnation would be to take a male immediately 

 after the transfer, cut through the pouch just back of the forward end 

 l)ehind the genital opening, and then examine the eggs in the hinder 

 part of the pouch for spermatozoa. This I had intended to do during 

 each of the past summers. Although there were numerous transfers 

 between fish kept in aquaria each summer, A'^et I saw the copulation 

 on one night only (in 1903) between two pairs of fish. The seeming 

 necessit}^ for keeping these fish for the early stages of segmentation 

 prevented my sacrificing either to determine this point. 



Huot (1902), Lilljeborg (1891), Ryder (1881), and others think that 

 the fertilization takes place at the time of copulation, while A. H. 

 Malm (1874) and Kroyer (1853) think that it follows later, and 

 Ekstroem (1831) believes it takes place while the eggs are in the pouch. 



