FOURTH REPORT OF THE DIRECTOR I907 6 1 



ing the climacteric period of the species. In a further chapter the 

 possible influence of the presence of spines on the development of 

 a retioloid structure in the periderm is investigated and the inference 

 attained that even the spinose forms of Diplograptus possess in 

 their thick periderm a layer of retioloid meshes, and of stronger 

 ledges, and that the development of this layer is roughly propor- 

 tional to that of the spines. This cause has combined with the 

 t'endency of the rhabclosomes to become lighter after the floating 

 and swimming habit had been adopted, and produced the order 

 Retiolitidae with a reticulate periderm. 



The dilatations, " disks," wings or vesicles of the nemacaulus of 

 Diplograptus, Climacograptus and Cryptograptus are separately con- 

 sidered and evidence brought forward to show that they were 

 inflations of the outer periderm of the nemacaulus through which 

 the virgula or axis passes uninterrupted. 



The verification by a recent investigation by Schepotieff, of a 

 former observation by the writer, that the axis of the sicula (the 

 virgula) passes through the nemacaulus and into the rhabdosome, 

 is discussed in a further chapter. Other chapters on the morphol- 

 ogy of the graptolites are devoted to the asymmetric section of the 

 rhabdosome in some graptolites (as Climacograptus typi- 

 cal i s), to the axes of the Dicranograptidae, the morphology of the 

 thecae of the Dichograptidae and Dicranograptidae. 



In a part entitled '' Notes on Phylogeny " the phylogenetic rela- 

 tions of the Leptograptidae and Dicranograptidae are first discussed 

 and the derivation of Dicranograptus from Dicellograptus shown. 

 It is argued that the branches of the Dicranograptidae formed 

 together always a more or less slender double spiral whereby certain 

 advantages of the arrangement of the thecae and a great elasticity 

 of the suspended rhabdosome were attained, but at the same time 

 the strain at the further (sicular) end where the two branches are 

 connected, increased ; hence the formation of the biserial portion in 

 Dicranograptus to strengthen this end. 



A synoptic and synonymic list of graptolites recorded from North 

 America concludes the general part of the memoir. 



A Devonic brittlestar 



In a iccent publication^ T called incidental attention to the dis- 

 covery by D. D. Luther of specimens of Ilelianthastcr in the Portage 

 (Cashaqua) shales at Interlaken, N. Y. a village lying on the divide 



* Report State r''dcoii1ologist iqoG, p. .36, 



