410 PBOP. &. GTLSON Oir SE&MEISI TALLY DISPOSED 



The secretion is not miscible with water, and presents the ap- 

 pearance of an oily fluid, though it is undoubtedly very different 

 from a fatty substance in th.e chemical sense of the term. 



These remarkable organs seem to deserve closer investigation 

 and minute description. Being engaged in other work, I have 

 asked one of my pupils, Dr. Henseval, to take up the subject. 

 He will shortly publish a paper dealing with these glands and 

 several others, as well as with the results of his researches on the 

 chemical nature of the *' oil " produced by the maxillary glands of 

 Gossus ligniperda *, a substance which seems to be identical with 

 that excreted by the thoracic glands of Trichoptera. 



A peculiar interest attaches to these thoracic glands of the 

 Larval Trichoptera, in its possible bearing on the question of 

 persistence of Annelidan features in the Tracheata. 



That they are newly acquired or adaptive organs, arising in 

 relation with the tubicolous habit, seems very unlikely, for if 

 the mere utility of their oily product is sufficient to account for 

 their appearance and development into important organs, there 

 seems to me no reason why they should be segmentally repeated. 

 One single gland, no matter where it lay, could furnish a suit- 

 able quantity of "oil" quite as well as the three moderately 

 large glands lying in close proximity to one another but on 

 separate segments. 



There is an organ undoubtedly homologous with the thoracic 

 glands of Trichoptera which has obviously nothing to do with 

 tubicolous life, i. e. the ventral gland, " Bauchdriise," described 

 by Professors Poulton and Schaff'er in certain non-tubiculous 

 caterpillars. 



It appears to me, therefore, much more probable that the 

 thoracic glands are inherited organs ; and that the aquatic and 

 tubicolous habits of the larva may account for their preservation. 



The question then presents itself, with which of the segmentally 

 disposed organs of Annelids and Peripatus are the thoracic glands 

 of Trichoptera to be considered homologous ? 



Only two kinds of organs may possibly be considered ancestral 

 to these glands — the nephridia and the coxal glands. If the 

 thoracic glands could be wholly or in part recognized as meso- 

 blastic in origin, little doubt would remain as to their nephridial 



* This paper was published during the passage of these pages through the 

 printers' hands, under the title " Etude compar^e des Glandes de Gilson," ' La 

 Cellule,' tome ix. pp. 329-354.— Ed, 



