BY R. J. TILLYARD. 



65 



occasional large amcEbocytes (Text-fig.9, ac) may also be found in 

 these positions, being probably caught and imprisoned during an 

 ecdysis, yet the travelling oat-shaped corpuscles or miocyte.s (mc) 

 are never found, in sections, anywhere in the gills except in the 

 blood-canals. A typical group of blood-corpuscles, including 

 both miocytes and amcebocytes, is shown, greatly enlarged, 

 collected near the wall of a blood-canal, at a point where it 

 impinges upon the hypodermis, in Text-fig.5. 



mc 



Text-fig.5. 

 Blood- corpuscles in blood-canal near base of gill of Austroarp-ion cyane 

 (Selys). From a transverse section { x 1000). Lettering as on p. 109. 

 The Aei'vous System. 



The Nerve-Supply of the Caudal Gills is, as might be ex- 

 pected, of very uniform structure throughout the whole Sub- 

 order, and appears to be but little affected by changes in the 

 form of the gill itself, or in the number of principal trachese 

 supplying it. The typical distribution of nerves to the gills, 

 here described, was found in all forms examined by me in the 

 families CalopterygidcB and Agrionidce. In the Lestidcf,, the 

 distribution is altered by the suppression of one pair of principal 

 nerves, in the median gill only. The nerve-supply of the lateral 

 gills remains the same in all three families. 



The last abdominal ganglion of the ventral nervous system of 

 the Dragonfly Larva is the large ganglion lying in the eighth 

 segment. This ganglion is formed by the fusion of three original 

 pairs of ganglia, viz., those of the eighth, ninth, and tenth 

 abdominal segments. It gives off four pairs of principal nerves, 

 viz., the genital nerves, the eighth segment nerves, the ninth 

 segment nerves, and the tenth segment nerves, the last being the 



