In 1909 I left the laboratory on July 3 and hence have no data for hatching this year other than that at this time the little catfishes were just beginning to burst their confining shells. However, the fuller data for 1910 show that on July 21 two lots of eggs were taken from a male; one with young embryos just beginning to show dark stippling, while the other lot had embryos with the body-walls cov ering about half of the yolk; these were from 2 to 3 weeks older than the former. See further on this point page 31, in section on the time of breeding. On July 30 an older fish measured 3.75 inches over all and had the edges of the body-wall united to make a raphe on the mid-ventral wall. The more advanced of the younger lot were able to skate on their yolk-sacs (see figure 11, plate iv), while others were not able to move these yolks which still anchored them to the floor of the aquarium.
All the evidence at hand indicates that the mortality in the ances tral mouth is far less than in even MacDonald hatching jars, and that the young brooded therein become free-swimming earlier. This is undoubtedly due to the fact that they feed while therein as pre viously described. Only once have I had a fully fledged young one from the mouth of the father. On July 20, 1908, my fishermen, men whom I know intimately and in whom I have great confidence, brought me a catfish about 4 inches long which they had gotten from the mouth of a male. It showed a mere seam on the flattened belly, the yolk being completely gone. No exact measurements could be made, since during the examination the fish jumped overboard and swam away like a flash.
From the data presented above it would seem that the time for the liberation of the young is from July 20 to 30, and under ordinary circumstances about the last week in July. Recalling that incubation begins May 20 to 30, we can approximate the length of the period of gestation as from 60 to 70 days. However, it is clear that the matter of the maximum or minimum limit will be largely if not wholly determined by the temperature of the water and the amount of small crustacea and other microscopic organisms contained therein and available as food for the larvae.
This may be summed up in one word—protection. Let it be recalled that these eggs are of enormous size (the average diameter of 327 eggs being 19.5 mm.) and that when in middle embryonic stages they are very attractive to the eye because of their blood-red vascular yolk investment. For these reasons, if laid as other fish eggs are, they could hardly be expected to escape the eyes of maraud ing fishes, but if any were so fortunate they would almost certainly be eaten by crabs, those scavengers from which practically nothing escapes. The result would be the inevitable extinction of the species. Moreover, there is yet another danger to which the ova are exposed. These catfish spawn and spend the hatching season on mud flats. If the eggs were discharged on such bottoms they would (because of their great weight, averaging 3.5 grams) sink into the mud and be smothered. To avoid these various dangers, these fish have to do one of two things to insure their perpetuity, i. e., to practice mouth gestation or to lay their eggs in nests which are guarded by one or both the parents. Some fresh-water catfishes have adopted the latter habit; the gaff-topsail has chosen the former.
The whole matter, barring his ignorance of the habit of buccal incubation, has been admirably put by Gurley (1902).
"It is almost impossible that a mud bottom should be a successful spawn ing-ground, as the eggs will almost inevitably be asphyxiated. Wherefore, fishes experiencing an impulse to spawn on such bottom will leave few de scendants to inherit their delicately sensitive mucous membrane, while those having an impulse to seek harder bottom will transmit to a larger progeny their more roborant mucous membrane. Further, the exception sustains the rule, the only species spawning on mud bottoms being certain catfishes, the males of which excavate nests, and attend to (probably aerate) the eggs, and care for the fry."