196 SUPPLEMENT TO THE 



regularly concavo-convex and without geniculation. As far as I am aware, Pander's 

 true type of Stro2)homena imbrex has not been hitherto discovered in our British Silurian 

 rocks. I therefore proposed for our Wenlock fossil the varietal designation of semighhosa^ 

 which could be made use of as a specific term should it be found that Pander's shell 

 was specifically distinct. The var. semiglohosa varies considerably in the degree of 

 convexity of its ventral valve. Some specimens form in profile an almost half circle ; 

 while others, on the contrary, are very gently or moderately concavo-convex, as is shown 

 by two fine examples collected by Mr. J. Gray, and now exhibited at the British 

 Museum. 



Another much smaller variety, and which never seems to attain one quarter of the 

 dimensions of the Wenlock fossil, has been found to occur abundantly in the Llan- 

 deilo, at Craighead and Balcletchie, in Ayrshire. We would propose to give to this 

 last-named variety the name of semiglohosina to distinguish it from the far larger Upper- 

 Silurian form. Its general shape, it is true, is much the same as that of the Upper- 

 Silurian variety, and it varies also considerably in the degree of convexity of its ventral 

 valve, also in the strength or coarseness of its thread-like radii. These rays are of almost 

 equal width, and in different specimens are separated by interspaces of greater or lesser 

 breadth, which are filled in with from three to five very fine longitudinal striae, with 

 here and there a shorter riblet. The interior of its dorsal valve, discovered by ]Mrs. R. 

 Gray, shows differences sufficient to authorise us in separating it from the Wenlock 

 form. The nature of the bottom of the sea may have had a good deal to do in relation 

 to the coarseness or fineness of the radii. Where the sea-bottom was composed of a 

 fine muddy substance, such as was the case at Craighead, the sculpture of the shell was 

 finer and more delicate ; while where the bottom was made up of a coarse material such 

 as at Balcletchie (conglomerate), the ribbing became coarser. Then there exists a very 

 considerable strati graphical vertical difference in the habitats of the two varieties. The 

 var. semiglobosina being found in the Llandeilo, the var. semiglohosa in the Wenlock. 

 All things considered, it seems desirable to retain for the two forms distinct varietal 

 designations. In PL XLI of my ' Silurian Monograph ' I figured the two varieties. 

 Pigs. 1 to 4 represent the variety semiglohosa ; figs. 5, 6, the variety semiglohosina. 

 In Scotland we have not hitherto discovered the Wenlock variety. 



