546 MATNARD M. METCALF, 



species (lasis hexagona and lasis costata-tilesii) ; the chambers of the 

 gland may be wholly wanting, while the ducts are present, the right 

 one being the larger (Salpa cylindrica) ; or the whole gland may be 

 absent {Thalia democraUca-mucronata and lasis cordiformis-^onaria). 



The presence of the gland in the solitary forms of all species 

 of SaJpidae and its absence in the chain forms of some species, recalls 

 the conditions in Boliolum where the gland is found in the nurse 

 (corresponding to the solitary Salpa) and in the sexual individuals 

 (corresponding to the chain Salpas) but is absent in the Phorozoid 

 of at least one species {DoUoJum affine). 



Peculiar ventro-lateral ganglia, outgrowths from the brain, are 

 found in the SaJpidae. There are two pairs of these, two large-celled 

 ganglia and two small-celled ganglia. These show even greater di- 

 versity than the gland. In the solitary forms, the pair of large-celled 

 ganglia is present in all species studied. The small-celled ganglia 

 are present in two species (Cyclosalpa pinnata and Salpa runcinata- 

 fusiformis) and absent in three species {Salpa cylindrica, Thalia and 

 lasis cordiformis-zonaria). In the chain forms both pairs of ganglia 

 may be present (the Cyclosalpas, Salpa africana-maxima and Salpa 

 runcinata-fusiformis) ; the small-celled pair may be wanting {Pegea) ; 

 or both pairs of ganglia may be absent {Salpa cylindrica, Thalia, 

 lasis hexagona, lasis costata-tilesii and perhaps lasis cordiformis- 

 zonaria). 



In the character both of the neural gland and of the ventro- 

 lateral ganglia the chain forms of the Salpidae show greater diversity 

 between species than do the solitary forms. I have found the same 

 to be true of the eyes (cf. Metcalf, 1893^). There is nothing in 

 the anatomy of the neural gland of the Salpidae to justify the guess 

 made by several authors that it may be a pair of otocysts ^), 



1) In so excellent a text book as Parker & Haswell's Zoology, 

 it seems strange to find this error repeated. They say (V. 2, p. 26): 

 "In Salpa there is an eye of a simple character and an otocyst" 

 (emphasis mine) "placed in close relation to the ganglion, in ad- 

 dition to eye-like bodies devoid of pigment. In Doliolum these are absent. 

 A sub-neural gland and duct are present in both these genera." The 

 reference to the accessory eyes of Salpa indicates that the authors 

 must have seen my paper on the eyes and neural gland in Salpa, for 

 they have not been described elsewhere. I do not see, therefore, how 

 they can call the two neural glands of Salpa "an otocyst" as Herdjian 

 has done. In referring to "a sub-neural gland and duct as present in 

 Salpa" they must be quoting Herdman's mistaken diagram of Salpa 



