126 Papers from the Department of Marine Biology. 
the females retain their eggs for a long time and that as these eggs 
gradually attain the size of pigeon eggs “the abdomen often swells 
up above the ordinary.’’ Hardly more definite is Wyman (1859), 
in writing of siluroids of Surinam: 
“The eggs become quite large before they leave the ovaries, and are ar- 
ranged in three zones corresponding to three successive broods, and probably 
to be discharged in three successive years: the mature eggs of a jarra-bakka, 
18 inches long, measure three-fourths of an inch in diameter; those of the 
second, one-fourth; and those of the third are very minute, about one- 
sixteenth of an inch.” 
Not so indefinite, however, is William Turner, the distinguished 
anatomist of Edinburgh, to whom Rev. Bancroft Boake had sent 
some specimens of Arius boakei, an oral gestator of Ceylon. Among 
these was one female, of which Turner writes (1867): 
“From the appearance of the abdomen it was evident that the ovaries 
were distended; and on opening into the cavity I found a sac-like ovary on 
each side of the middle line. Each ovary measured 21% inches in length, 
and extended forward almost as far as the pectoral fin, where it formed a 
rounded free end, whilst posteriorly it was somewhat constricted, and opened 
by an orifice common to it and its fellow immediately behind the anus. The 
ovisac contained a very large number of eggs in various stages of growth. 
Some were like minute granules, others (and these very numerous) like 
medium-sized shot, whilst a third set equaled in size grapes, or small cherries, 
and very materially exceeded therefore the size usually attained by the eggs of 
osseous fishes. These last, only six in number in each ovary, had evidently 
almost reached the full period of intra-ovarian growth. Each ovum was 
attached to the inner wall of the ovisac by an independent pedicel, the atrophy 
of which would necessarily precede the discharge of the egg.” 
Equally definite is Francis Day (1873), who gives us an account of 
the structure of the ovary of certain Indian siluroids. In describing 
the habits of a number of well-known marine catfishes practicing oral 
incubation, he says: 
“Next, the females came under observation. On tracing up the ovisacs 
it appeared that a very large number of eggs existed in them, but no¢ all of 
the same size [italics Day’s]. On the part farthest removed from the outlet 
the eggs were of full size (about half an inch in diameter), and about 50 in 
number; whilst other batches of much smaller size existed, evidently to take 
the place in due time of the larger ones when they had been deposited. The 
full-sized eggs were each attached to the inside of the ovisac by a pedicel of 
varying length, distinctly supplied with blood-vessels of considerable size.” 
The next notice of the ovaries of an Indian catfish is an incidental 
one from Edgar Thurston in 1900. On the Malabar coast of south- 
west India he dissected an Arius (species undetermined) and found 
that what he calls the ‘double uterine cavities’ contained respectively 
56 and 75 eggs of about 13 mm. in diameter. 
At the 1907 meeting of the French Association for the Advance- 
ment of Science, Jacques Pellegrin read a paper on buccal incubation 
