[168] The reverse nomenclature to this is rather inconveniently employed by Metschnikoff.
[169] This has not been shewn in the case of Hydrophilus.
[170] According to Kowalevsky the history of the dorsal plate is somewhat different. He believes that on the absorption of the amnion the ventral plate unites with the serous membrane, and that the latter directly gives rise to the dorsal integument, while the thickened part of it becomes involuted to form the dorsal tube already described.
[171] Tichomiroff (No. [420]) denies the existence of a true invagination to form the mesoblast, and also asserts that a separation of mesoblast cells from the epiblast can take place at other parts besides the median ventral line.
[172] If these structures are equivalent to appendages, they may correspond to one of the pairs of antennæ of Crustacea. From a figure by Fritz Müller of the larva of Calotermes (Jenaische Zeit. Vol. XI. pl. 11, fig. 12) it would appear that they lie in front of the true antennæ, and would therefore on the above hypothesis correspond to the first pair of antennæ of Crustacea. Bütschli (No. [405]) describes in the Bee a pair of prominences immediately in front of the mandibles which eventually unite to form a kind of underlip; they in some ways resemble true appendages.
[173] In Smynthurus, one of the Collembola, there are, according to Lubbock, only two stigmata, which are placed on the head.
[174] Graber’s view on this point may probably be explained by supposing that he has mistaken a passage of yolk cells into the blastoderm for a passage of blastoderm cells into the yolk. The former occurrence takes place, as I have found, largely in Spiders, and probably therefore also occurs in Insects.
[175] This point requires further observation.
[176] For a systematic account of this subject the reader is referred to Lubbock (No. [420]) and to Graber (No. [411]). He will find in Weismann (Nos. 430 and 431) a detailed account of the internal changes which take place.
[177] Brauer and Lubbock (No. [421]) have pointed out the primitive characters of these forms, especially of Campodea.