138 



ME. R. J. TILLYARD ON THE RECTAL BREATHING-APPARATUS 



Double Embedding. — The ordinaiy single embedding method yields very 

 satisfactory results in nearly all cases. For the papillate types, and when- 

 ever a "very exhaustive histological analysis is required, celloidin embedding 

 may be carried out from the pure cedar-oil stage as follows : — The object is 

 embedded in celloidin in the usual manner, but with 1 part of cedar-oil 

 added to 4 parts of the celloidin solution. On reaching the strongest celloidin 

 solution the proportion of cedar-oil is reduced to 1 part in 5. The hardening 

 is done in 5 parts of chloroform to 1 part of cedar-oil^ the mixture being 

 frequently changed. The block must then be embedded in paraffin dissolved 

 in benzol to concentration in a closed vessel. The benzol must be allowed 

 to evaporate very slowly for a week or more, until the paraffin is quite dry 

 and firm. Sections may then be cut from the block as fine as 3 //. in 

 thickness. The method is, however, not only a very protracted one, but 

 offers considerable difficulties, especially in the successful elimination of the. 

 benzol. 



Material Studied. 



Nineteen species, belonghig to fifteen genera, have been studied for the 

 purpose of this paper. Microphotographs were taken of the gills of each of 

 the nineteen species^ and sets of serial sections prepared from the rectum 

 of .every genus represented. In some genera sections were obtained not only 

 from the more advanced larval instars, but also from the earlier stages (first 

 to eighth instars). The total number of sets of serial sections prepared and 

 studied amounted to over seventy. 



In the following table I have arranged the material studied in the accepted 

 systematic order, giving family, subfamily, and tribe, together with columns 

 indicating the system and type of gill to which each genus belongs : — 



