BRITISH SILURIAN BRACHIOPODA. Ill 



same valve with age becomes gradually more and more convex or gibbous, and loses 

 by degrees all trace of the longitudinal depression. 



The front line is also either straight or more or less curved upwards, so much so that 

 many specimens show in the dorsal valve a well-developed mesial fold, with a corre- 

 sponding sinus in the ventral one. The ribbing varies also to a very considerable extent 

 in different specimens. In young individuals the ribs are few in number, and in this 

 condition it approaches in shape and character to similar-sized examples of Atrypa 

 Barrandei. The number of ribs seems also to increase rapidly with age. Some specimens 

 with very convex or gibbous dorsal valves are covered with numerous fine radiating ribs, 

 while others of the same size show a much smaller number, and these more close and pro- 

 minent. The intersecting concentric lines or ridges due to growth are also much stronger, 

 closer, or more widely apart in some individuals than in others, still these varied 

 individuals are linked one to the other by gradual passages. Feeling anxious to acertain 

 whether there existed interiorly any gradual increase in the number of spiral coils from the 

 young up to the adult condition, I placed in the hands of the Rev. Norman Glass a number 

 of well-preserved specimens at different stages of growth, and some of these he developed 

 with his usual ability, and was soon able to show, and in the most distinct manner, that the 

 number of coils increased with the growth of the shell. In a specimen measuring four lines 

 In length and breadth there were only five convolutions in each spiral cone, and these spe- 

 cimens much resemble those of Atrypa Barrandei, and in all probability, if not certainly, in 

 still younger specimens, Mr. Glass would have found not more than three or four coils. 



In a specimen measuring five lines in length he found six coils ; in another six 

 lines in length seven coils, and so on ; no doubt the increase would proceed up to 

 sixteen convolutions in each spiral, the usual number found in full-grown specimens. 

 Mr. Glass showed also that the spiral appendages in full-grown Dudley specimens 

 commence by forming very large coils, that these are succeeded by others smaller in 

 diameter, forming two comparatively narrow cones, that the extremities of these 

 cones are closer together in some specimens than in others, and that sometimes one spiral 

 cone seems a little longer than the other, one cone, for example, showing only fourteen 

 convolutions, whilst its companion cone has as many as sixteen. Mr. Glass ascertained 

 likewise that the basis of the spiral cones in young specimens with flattened dorsal valves 

 is not level, the two inner sides of the principal coils being slightly higher up than the 

 two outer sides and turned towards the margin of the shell, which is exactly what we 

 find to be the case, and have represented, in Atrypa marginalis and Atrypa Barrandei, the 

 dorsal valves of which are also nearly flat. As the shell grows and the dorsal valve 

 becomes more convex, the basis of the spirals becomes more level and the spiral cones 

 more elevated, as we have figured them in our plate. Sup., PI. VII, figs. 4 to 6. The 

 principle of variation of shape in the spirals of certain genera and species of Atrypida 

 seems to be the providing of such a form of spirals as should allow the greatest length 

 of coil possible in the interior of the shell ; for example, in Glassia obovata and G. 



16 



