CARADOCIAN CYSTIDEA FROM GIRVAN. 497 



continuity broken, with a sea-shelf lying to the north of it and another to the south " 

 (Weller, 1898, p. 702 ; and compare the maps of Schuchert, 1910). 



I 567. Turning more particularly to Scotland, we read that " the sequence of 

 Caradoc rocks in the Girvan region points to the existence of an extensive land-surface 

 to the north of the Silurian [Ordovician] sea. The strata represent shore-deposits 

 which were accumulated on a gradually subsiding area. But here, as elsewhere, 

 the close of this period is marked by a great change in the life-history of the 

 system" (Peach and Horne, 1899, p. 55). Similarly in the Northern Belt "the 

 transport of coarse terrigenous material from the north-west extends further south 

 with each successive period" (p. 53). So also, " while oceanic conditions of deposition 

 [still] prevailed over a large part of the Central Belt in Llandovery time, the coarse 

 terrigenous materials were carried further south, till in Tarannon time they overspread 

 the sea-floor of the whole central region" (p. 56). Later conclusions, explanatory of 

 the physical changes that facilitated the immigration from America, are summarised in 

 the final paragraph of Dr Peach's address to the Geological Section of the British 

 Association (1912), delivered after the manuscript of this memoir had been completed. 



§ 568. The route to Canada is made clear to us by the preceding observations. 

 That to the Baltic was probably somewhat sinuous and liable to interruption. That to 

 Bohemia is less clearly defined, and probably was not continuous for more than a short 

 period ; the evidence of the trilobites shows that it existed during the deposition of 

 the preceding Whitehouse Group, possibly to a slight extent even earlier, but that the 

 connection then ceased suddenly. It is natural to suppose that the trilobites migrated 

 more rapidly than the stalked Echinoderms, and that the latter underwent more change 

 on the way ; therefore it is not necessary to imagine a continuous shore-line from 

 Girvan to Bohemia while the Drummuck Group was forming. 



In brief, then, we seem to have alighted on a highway skirting the Atlantic Basin, 

 along which forms are slowly migrating from east to west and from west to east, and 

 meeting on the way, and being modified as they pass. 



II. Anatomy. 



The following is a brief summary : — 

 Heterostelea. 



§ 569. Rhipidocystis. Details are added to the very brief description by Jaekel 

 (§§ 34-36). 



§ 570. Dendrocystis. The theca is shown to have a more definite form than previ- 

 ously recognised, tending to a reversed heart shape (§ 45). The discovery of undoubted 

 anal structures in D. scotica permits the orientation of the theca (§§ 73, 141). Modifi- 

 cations of the thecal plates near the attachment of the stem and brachiole respectively 

 are described (§§ 59-61). The structure of the thecal stereora is now first noted (§ 53). 

 A more complete account of the appendage confirms its brachiolar nature (§ 63). The 



