Manchester Memoirs, Vol. Ixii. ( 191 7) 127 



for identification, but the description shows tfte form to be a 

 variety of R. soldanii, and probably identical with d'Orbigny's 

 species R. umbilicata. Halkyard's separation of R. umbilicata 

 in the Biarritz material is more than justified owing - to the very 

 strong local development of the typical R. soldanii.) 



320. Rotalia calcar, (d'Orbigny). 



Calcarina calcar, d'Orbigny, 1826, Ann. Sci. Nat. vol. VII, p. 



276, No. 1. 

 Rotalia calcar, Brady, 1884, Chall. Rep., vol. IX, p. 709. 



pi. CVIII, fig. 3; and fig. 4? 



Frequent. Examples small as a rule; the larger ones lose 

 a good deal of the clear feature of the species, the exterior 

 margin of the segments becoming curved instead of pointed. 



(The species show gradual transition from the long spined 

 R. calcar to spineless forms very near R. venusta, Brady.) 



321. [Nonionina kochi, (Hantken.)] 

 321 a. Rotalia pulchella, (d'Orbigny). 



Calcarina pulchella, d'Orbigny. 1839, in De la Sagra's Hist. 



Phisiq. etc., de Cuba, pt. "Foraminiferes," p. 80. -1. V, 



figs .16-18. 

 Rotalia pulchella, Brady, 1884, Chall. Rep. vol. IX, p. 710, pi. 



CXV, fig. 8, a, b. 



Rare. The few specimens found are small and weak, and 

 resemble d'Orbigny's rather than Brady's figure. 



(A close examination of Halkyard's specimens makes it 

 impossible to retain the species under his determination. They 

 cannot be compared either with d'Orbigny's original Cuba 

 figure, or with the much more familiar Malay type assigned by 

 Brady to d'Orbigny's species although very dissimilar from 

 d'Orbigny's figure. Halkyard's specimens instead of being 

 rotaliform in structure, as in the d'Orbigny-Brady types, are 

 involute, and must therefore be assigned either to the genus 

 Nonionina, or to its globigerine isomorph Pullenia. Apart 

 from this generic distinction the arrangement of the spines 

 is quite different to the d'Orbigny-Brady types, in both of which 

 the number of spines is limited to three and they are soiid 

 and set at equally divergent angles from the centre of the test. 

 In the Biarritz specimens, which are unfoitunately few in num- 

 ber and mostly damaged, the spines are tubular and their number 



