248 NATURAL SCIENCE. October. 



has already been shown by the note of interrogation always placed 

 after the name Dendrograptus, I am uncertain to what genus the 

 forms ought to be referred. This depends on the fact that the old 

 generic diagnoses were drawn up at a time when the structure of 

 the dendroids was as good as unknown ; they are therefore 

 generally of no, or at least of undecided, value, and always taken from 

 the external characters of specimens found in slate, and usually in a 

 bad state of preservation. The classification therefore needs a 

 thorough revision, but the time does not seem to me yet ripe for such 

 an undertaking, since we are not yet in a position to estimate the 

 relative value even of characters collected from the internal structure. 

 For instance, a classification of the Dendroidea according to the way 

 the gonangia open might perhaps be as artificial as the old one. At 

 all events, we had better postpone the task of classifying till more is 

 known about the structure of the proximal end. 



Thus, forms with a sicula, such as Dictyonema flabellifovme, and forms 

 without a sicula, such as Dictyoncma peltatuni, are now put together, on 

 account of the connecting fibres between the branches. I do not yet 

 know the structure of the proximal end in more than the last-named 

 Didyonema and one still undescribed form. 



In Didyonema peltatuni a large number of branches spread centri- 

 fugally within a disc, and then rise up at different distances from the 

 middle of the disc, branch, anastomose, and join again by means of 

 the ordinary connecting fibres, thus giving to the rhabdosome the 

 ordinary funnel-shape. The proximal ends of the branches do not at 

 all possess the intricate structure that characterises the distal parts, 

 but in cross-section look just like a branch of Monograptus, except that 

 the virgula is missing. In the region where this simple structure 

 passes into the more complicated one, the rhabdosome was broken 

 when cleaned out from the flint. 



The other undescribed species begins with a disc, from which a 

 stem proceeds, much as in Odontocaulis keepingi, Lapw. A section 

 through the lowest part of this stem also resembles the cross-section 

 of a Monograptus. There is no virgula. The periderm is very thick. 

 The one hole is larger, and is the section of a theca that opens on the 

 stem. The other one is small, and very soon proves to be a budding 

 individual, in that it, some sections higher up, has produced and con- 

 tains the ordinary three individuals : a theca, a budding individual, 

 and a gonangium. Probably the course of growth is the same in each 

 separate branch of Didyonema peltatuvi. 



There is no reason for fitting in the graptolites in any of the 

 classes of the animal kingdom that now exist. The resemblances ta 

 Rhabdopleura, the bryozoans, the hydroids, and others, are only such as 

 easily arise between colony-building animals of even a widely-separated 

 systematic position. Even as to their approximate position in the 

 system not much more can be said than that they are Invertebrata ; 



