Rotifera of some Pacific Islands. By J. Murray. 433 



lateral spines three on each side ; second dorsal series of four 

 spines ; two spines on rump. 



Variety (figs. 2a-2c). — Spines on anterior border of trunk all 

 small. Main dorsal series of ten spines; lateral spines four on 

 each side ; second dorsal series of two spines, widely separated ; two 

 spines on rump ; a transverse ridge on first foot -joint. 



The Hawaiian varieties are chiefly remarkable for the increase 

 in the number of lateral spines — no other variety is known with 

 more than two on each side. There is a variety in Scotland which 

 has the second dorsal series of two spines widely separated, as in 

 the second Hawaiian variety. 



Callidina punctata Murray. 



A rare species, apparently confined to warm countries. Dis- 

 covered in Africa in 1910, and now known in Australia and 

 Hawaii. 



Callidina multispinosa (Thomp.). Plate XV. fig. 5. 



Variety. — Having long setse, short spicules, knobs, and pectinate 

 processes. On each side five long setse on the anterior trunk. In 

 the angle between the first and second trunk-segments a short 

 acuminate process. Transverse dorsal row, at widest part of 

 trunk, of small spicules, but the processes at the ends of the row 

 broad and pectinate or dentate. Processes at posterior angles of 

 trunk and on rump thick conical knobs. On the foot a pair of 

 lateral setae, with bulbose bases, and a transverse series of small 

 truncate or bifid processes. 



The chief peculiarity is the pectinate processes on the central 

 trunk. A short-spined variety from Fiji has similar processes on 

 the anterior trunk. 



Rotifer longirostris (Janson) var. fimbriate/, Murray. 

 Plate XV. fig. 7. 



The Hawaiian form is remarkable for the great elongation of the 

 processes, which suggested the name fimbriata. They are longest 

 at the widest part of each segment, and there attain to a length of 

 10 or 12 fi. 



Had this extreme form been first discovered, no doubt it would 

 have been regarded as a distinct species. The firm sharp processes 

 might not have been recognized as the product of secretion. 



These rod-like projections appear to be of the same nature as 

 those secreted by Dissotrocha macrostyla and some other species, but 

 they show no tendency to cohere to form tubeicles. This is probably 

 because the secretion is of a kind which hardens more quickly. 



The strongest arguments for this form being specifically distinct 

 from R. longirostris are found in the decidedly more nodose trunk, 



