Page:Proceedings of the Royal Society of London Vol 69.djvu/24

16 These tables show that, whilst the serial number of the girdle- piercing nerve is identical, the average number of collector branches is slightly less in the embryos than in the adults.

Consequently, for practical purposes, such embryos may be considered as being of equal values with the adults. From analogy with the case of Mustelus quoted above, it might have been expected that these older embryos would have shown a slightly larger number of collector nerves than the adults. It must, however, be borne in mind that these embryos of Acanthias are rather more advanced than those of Mastdus, which were previously considered. As far as I have been able to observe, an Acanthia* embryo of a given length about this stage is more advanced than a Mustelu* embryo of the same length.

Returning now to Table VIII, if we are to explain the correlation there expressed on the side-fold excalation theory, we must suppose that the lesser number of collector branches associated with a more rostral position of the girdle is due to excalation of vertebrae through which pass nerves to the collector as well as vertebrae in the pre- collector area. This will be made clearer by the consideration of a concrete case. In fig. 6 two cases have been selected, one in which

FirsG

b.

Coiiector &re& shaded.

FIG. 6.

the first girdle-piercing nerve is thirty-five, and the other in which it is forty. From Table VIII. we learn that the average number of collector nerves associated with these two positions of the girdle are ten and twelve respectively. Now if we are to derive the case in which the girdle is more rostrally situated from that in which its position is more caudal, the figure shows that we must assume two segments to have been excalated in the collector area, and three in the pre-collector portion of the trunk. The assumptions made are (1) that the girdle-piercing nerve is homologous in each case, and (2) that a more rostral position of the girdle is in itself no argument for a lessened collector area. Consequently we must assume that on the side-fold excalation theory, excalation takes place both in the collector and the pre-collector areas.