514 Scientific Proceedings, Royal Dublin Society. 



of the occurreuce of stems with foliage leaves' attached. It is thus evident 

 that the leaves were deciduous at an unusually early age of the slioot. 

 Fortunately the specimen already mentioned, and named by Brongniart 

 Lepidodendron Griffitliii, and other specimens (e.g. Plate XXXV, fig. 1) furnish 

 the missing evidence, and show how distinct the foliage of B. kilto7-kense is from 

 that of other species, such as B. 2»inctatum and B. minnlifolium, in which 

 attached, more or less lanceolate, Lycopodium-like leaves have long been 

 known. The foliage leaf of B. lUforkeiise, as Plate XXXVI, fig. 1, shows, occurs 

 singly in horizontal or obliquely ascending whorls of 10-20 members, and 

 forms apical tufts. It is a linear-subulate leaf, 12-15 cm. long, with a 

 single central vascular bundle running throughout its length. It tapers 

 gradually from the point of attachment, where it is 1 mm. wide, to its apex, 

 and shows no specialized basal portion. It is, on the whole, such a leaf as 

 one would expect, judging from the characters of the fertile leaf, and 

 assuming that in such a primitive type of plant the fertile and steriler leaves 

 would differ but little from one another in general form. 



The fact that B. hiltovhense is one of the earliest of the Lepidophytes 

 seems to indicate that the small lanceolate leaf of B. punctatum, a much 

 later species, is derived from the long narrow leaf of B. Idltorkense, and 

 persists to-day in the genus Lycopodium. A Lepidodendron with leaves a 

 metre long has been recorded (25). If the leaf is a mere appendicular 

 outgrowth of the axis, arising at first as a slight proliferation of it, 

 B. hiUorkense should show, one would expect, a small inconspicuous leaf 

 instead of the pronounced elongated one described. It is worthy of note 

 that the leaves of Cyclodigma hercynium of the Harz Mountains, as shown, 

 have the same shape as those of B. killorkense, but are only one-half or one- 

 third of their length. But for this, I sliould be inclined to agree with 

 Nathorst, who sees nothing to distinguish Potonie's G. hercynium from 

 B. killorkense. It is possible that the leaves of C. hercynium, figured by 

 Potonie, have their free ends buried in the stone (op. cit., p. 39, fig. 16), 

 and in consequence do not show their full length. 



Stigmaria Stage. 



One of the commonest Carboniferous fossils is Stigmaria ficoides, now 

 known to be the rhizome or underground stem of Sigillaria and also of 

 Lepidodendron. As Bothrodendron is so closely allied to these two genera, 

 one might naturally expect it to have a more or less similar Stigmaria- 

 stage. Is there any evidence of this ? 



' In Kidston's pithy account of B. kiltorhense (" Guide," 1886) leayes are described; whether 

 found attached or not is not stated. 



