SOME FERNS FROM THE CARBONIFEROUS FORMATION. 153 



angle ; those angles formed by subsequent dichotomies are a little more acute. 

 The pinnae of the barren fronds possess a broad foliaceous expansion cut into 

 spirally bent lobes, in which the nerves are indicated by dichotomously 

 dividing ridges. Fertile pinnae dividing dichotomously and reduced to winged 

 circinately convoluted rachis-like segments. The convolutions of the basal 

 portion of the pinnae overlap each other; their ultimate divisions are narrower, 

 less prominently winged, and do not apparently overlap each other, or only 

 do so to a limited extent. 



Remarks. — Specimens of this fern have been in the Collection of the Geolo- 

 gical Survey of Scotland for many years, that figured on Plate VIII. fig. 13, 

 having been collected by the late Mr Richard Gibbs about twenty-live years 

 ago. The portions of the species with which I first met were fragments of the 

 scorpiod fruiting pinnae. These, I thought, might perhaps belong to Triphyl- 

 lopteris Collombi, Schimper,* or to Cyclopteris {Aneimites) Aoadica, Dawson,f 

 both of which species have very close affinities to each other, if not specifically 

 identical. 



Associated with the British examples, though careful examination was 

 made, no barren pinnules were ever discovered that could be identified with 

 either Schimper's or Dawson's ferns. 



It was only towards the end of 1884 that my difficulties in the identification 

 of this fern were removed by Mr John Rhodes, Fossil Collector to the Geolo- 

 gical Survey of England, finding the specimens figured on Plate VIII. figs. 11, 12, 

 which show the barren condition of this plant. I had previously seen a small 

 fragment of the barren condition of AUicornopteris convoluta from Docken 

 Beck, Eskdale, collected by Mr A. Macconochie, one of the Fossil Collectors 

 to the Geological Survey of Scotland, but had erroneously identified it as 

 Rhacopliyllum Lactuca,\ to which small fragments have a great resemblance, 

 so much is this the case, that with fragmentary specimens it is almost impossible 

 to distinguish them. The fruiting portions of A. convoluta have apparently a 

 more strongly winged rachis than occurs in Cyclopteris Acadica, or in Triphyl- 

 lopteris Collombi; but here also small fragments of the ultimate segments of 

 the fruiting portions of AUicornopteris convoluta would be with difficulty 

 distinguished from fragments of the fruiting portions of the two ferns already 

 mentioned (see Plate VIII. fig. 15). 



* Triyhyllopteris Collombi, Schinaper-Zittel, Handbuch der paleontologie, ii. Band, 1 Lief, p. 114, 

 fig. 84, 1879; Traite d. paleont. veget., vol. i. p. 479, pi. cvii. fig. 13; "Les veget. foss. du terrain de 

 trans, d. Vosges" (in Le terrain de trans, d. Vosges, by J. Koechlin-Schluniberger and "W. Ph. 

 Schimper, Strasburg, 1862), p. 339, pi. xxvii. figs. 8-11 (Sphenopteris). 



f Cyclopteris (Aneimites) Acadica, Dawson, "Geological Survey of Canada," Fossil Plants of the 

 Lower Carboniferous and Millstone Gh'it Form, of Canada, p. 26, pL vii. figs. 53-63, 1873 ; Acadian 

 Geol, 2nd ed., p. 481, fig. 75, 1868; Quart. Jour. Geol. Soc, vol. xxii. p. 153, pi. viii. fig. 32, 1865. 



+ Trans. Roy. Soc. Edin., vol. xxx. p. 540. 



VOL. XXXIII. PART I. U 



