504 E. V. Cowdry, The relations of mitochondria and other cytoplasmic etc. 



blackened network. Section from which fig. 27 was drawn was 6 /< in thickness 

 (pag. 492). 



28, 29 and 80. From a single spinal ganglion fixed in Bensley's formalin- 

 bichromate-sublimate mixture (Bensley, 1910). Canalicular system as clear spaces 

 (pag. 494). 



Plate XVII. 



These figures have been drawn in the same style as those of the preceding 

 plate. The first row logically belongs with plate XVI, for in it three characteristic 

 types of canalicular formation are shown. The second and third rows of figures 

 show gradations in the completeness of the staining of the canals, and the last row 

 is designed to illustrate variations in their form. 



31, 32 and 33. From a spinal ganglion prepared by Kingsbury's 1911 modi- 

 fication of the Weigert hematoxylin method. The canalicular apparatus, in the 

 original preparation, was stained a dark brownish black color. The three types of 

 configuration are shown (pag. 494). 



34, 35 and 36. From the same section of a single spinal ganglion fixed in 

 trichlorlactic acid and stained with Weigert's resorcin-fuchsin (Holmgren, 1901). A 

 gradation in the completeness of the coloration of the canalicular apparatus is 

 evident (pag. 492). 



37, 38 and 39. From the same section of a single spinal ganglion treated 

 with a 2^Jq solution of osmic acid for eight days (Kopsch, 1902). They show, like- 

 wise, a gradation in the blackening of the canalicular apparatus (pag. 494). 



40. Fixed in Bensley's formalin-bichromate-sublimate mixture and stained 

 with iron hematoxylin. The clear, unstained canalicular apparatus is broken up 

 into island-like, disconnected fragments (pag. 494 and 495). 



41 and 42. Neighbouring cells from the same section of the same spinal 

 ganglion fixed in Bensley's formalin-bichromate-sublimate and stained with iron 

 hematoxylin. They illustrate the variability in the size and in the extent of the 

 canalicular apparatus, and a comparision of the latter with fig. 28, which was 

 drawn from a cell from the same section, shows considerable variation in the dia- 

 meter of the canals. Neither of the two are extremes (pag. 495). 



