HISTORY OF RESEARCH. xlix 



III. 



Slender cylindrical brandies, with tubular cellules arranged in single (or in 



double) series. Cellules not in contact in any part of their 

 Rastrites. J l 



lengtli. 



IV. 



Species having a common axis or rachis, with slender lateral alternating 

 Thamnograptus. branchlets. Cellules unknown. 



V. 



Species having a common axis, more or less frequently bifurcating, with 

 pinnulge closely and alternately arranged on the opposite 

 sides ; cell-apertures on one face of the pinnulae. 



VI. 



A simple flexuous rachis, with slender flattened pinnulae, arranged in alternating 



„ ±1 ± order at close and regular intervals on the two sides. Cell- 



Buttiograptus. 



apertures unknown or circular. 



VII. 



Strong stems, which are numerously branched. Branches and branchlets 

 Oldhamia. slender, arranged in whorls. Cellules undetermined. 



Nomenclature. — It will be seen from the above synopsis that Hall still maintains 

 his original view that all the simple stiped forms described by Barrande and others 

 are really only isolated branches of a more complex form, and he therefore rejects 

 all such generic names as Monograpsus, Didymo-, Tetra-, and Dichograpsas. He 

 writes, " These subdivisions may be of some value when the entire frond and all 

 its appendages are preserved, but unfortunately this is rarely so ; and when we 

 have but fragments of the stipes or branches there is no force or value in the 

 application of these terms ; we are thus reduced to the necessity of adopting the 

 old term Graptolithus." In criticising M'Coy's genus Didymograpsus he points out 

 that the name had been used for two distinct groups, namely, those of the type of 

 Graptolithus patulus, or the true Didymograpti (as we know them at the present 

 day), and those of the type of Gr. divaricatus (Dicellograpti), and he shows how 

 valueless are the genera founded on the number of stipes by citing the east' of 

 Didymog. caduceus (Salter), which he believes to be a four-stiped form, though it so 

 closely simulates one with two stipes. 



Hall points out that the corneous disc in Graptolites is not a character of 

 generic value, as it occurs in some 1-, 8-, and 10-stiped forms, while other species 



