CARADOCIAN CYSTIDEA FROM GIRVAN. 495 



genus, Dendrocystis, is " not found elsewhere in the British Isles, though occurring in 

 other parts of the world." " Thirdly, there are . . . generic constituents which, though 

 not unknown from other Lower Palaeozoic regions in the British Isles, are decidedly 

 uncommon and rare; such are" Pleuvocystis and Cheirocrinus. "We may further 

 note the unusual stratigraphical horizon " of Dendrocystis, previously, if our correlations 

 are correct, unknown above the base of the Caradocian. " The occurrence of some 

 forms identical with foreign and not British species " is paralleled by Cheirocrinus 

 interruptus, " while others are less allied to members of English [British] Palaeozoic 

 faunas than to members of continental faunas." The adaptation of Mr Reed's remarks 

 might be continued, but enough has been said to show that, however peculiar this 

 Cystid fauna may be, its peculiarities are shared by the Girvan fauna as a whole. 



§ 561. To come to details. Dendrocystis has hitherto been recorded only from 

 Bohemia and Russia (§§ 41-43). Now the oldest species is reported from the 

 Tremadocian of Herault (§ 43). Then follows D. Barrandei from the base of the 

 Llandeilian (d2) in Bohemia. The genus is next found in the Baltic provinces, 

 and again at the base of the Caradocian (d3, d4) in Bohemia. The route of migra- 

 tion to Scotland was probably not through Russia, for D. rossica presents peculiar 

 characters, and D. scotica is far nearer the Bohemian species ; indeed, it resembles the 

 older D. Barrandei almost more than it does D. Sedgwicki, but this is partly due to 

 the mode of preservation. 



§ 562. Meanwhile Dendrocystis had probably found its way to Canada, where 

 D. C?) paradoxica occurs in the Trenton Limestone (§§ 42, 155) ; and it is also interest- 

 ing to note that the allied genus Rhipidocystis may have a North American representa- 

 tive in the fossil called Platycystis Faheri (§ 37). These facts, however, are not enough 

 to prove any direct connection between North America and the Baltic. 



§ 563. Cothurnocystis (§ 161), being not merely a new genus, but presenting a plan 

 of structure hitherto unimagined among Echinoderms, might be thought to have no 

 bearing on these questions. I have, however, attempted to prove that its nearest allies 

 are Ceratocystis and Trochocystis (§§ 245-260), of which the former is as yet known only 

 from Bohemia, and the latter only from Bohemia and SW. Europe, in all cases so long 

 before as the Middle Cambrian. 



§ 564. Turning to Cheirocrinus, we find quite other connections. It is true that 

 the genus has previously been recorded from Great Britain (§ 309), but the evidence, so 

 far as I can ascertain, has always been in the form of scattered plates not specifically 

 determinable. These appear to be generally of Llandeilian age. Plates, probably of 

 this genus, also occur in the Balclatchie Group of Girvan (§ 4), which has been con- 

 sidered Llandeilian but is now transferred to the Caradocian. I find no record of the 

 genus in rocks of Ashgillian age here or elsewhere. 



It follows that the Cheirocrinus of the Starfish Bed might conceivably have been 

 descended from Scottish, or at least British, ancestors. Or if not, the genus was 

 so widely spread that representatives might have wandered into Girvan from any 



TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 6). 65 



