1854.] Linnean Society. 353 



the Palms almost all fossil plants presenting strange and anomalous 

 forms. This remark applies with full force to Sigillaria and Lepi- 

 dodendron, genera belonging to the Palaeozoic formation ; and 

 startling as it may be, it cannot be very surprising to us, when we 

 consider that even in our own times notions equally erroneous 

 prevail to an almost incredible extent. For instance, a principal 

 share in the formation of coal is still ascribed by the generality of 

 geologists to the Ferns, although it has been proved by me, by a 

 series of careful observations on numerous carboniferous formations, 

 that such can be claimed only for the Sigillarias and Stigmarias ; that 

 a subordinate share only is due to the Araucarias and Calamites, con- 

 tained in the anthracite coal ; and a still less significant portion to 

 the Lepidodendrons, the Ferns, and the remaining members of the 

 flora of the Carboniferous period. 



" A diligent study of the extinct flora has demonstrated that the 

 Palms occur more rarely than was thought in the transition rocks 

 and the Carboniferous formation, and more frequently than was 

 believed in more recent formations. Generally speaking, we know 

 at present only trunks, leaves (both fan-shaped and pinnatisect), 

 and a few fruits of fossil Palms ; flowers have as yet not been 

 discovered, and the spathes collected have hitherto not been satis- 

 factorily proved to belong to the natural order under consideration. 

 These fragments, remnants of members of former creations, have 

 been distributed under nine genera and seventy -eight species. For 

 the trunks, there have been adopted the genera Palmacites, Brongn., 

 and Fasciculites ; for the leaves, Flabellaria, Sternb., Zeugophyllites, 

 Brongn., Phcenicites, Brongn., and Amesoneuron, Gopp. ; for the 

 spathes, Palceospatha, Unger (founded, as has been stated, upon 

 doubtful materials) ; for the fruits, Baccites, Zenk, and Castellinia, 

 Massal. The genus Burtinia of Endlicher, placed by Unger among 

 Palms, belongs without a doubt to PandanecB, and is identical with 

 Nipadites. 



" Although our present imperfect knowledge of fossil plants 

 renders it an almost useless task to speak of the proportion which 

 the number of Palms bears to that of the other members of the 

 extinct flora, yet it may be remarked that, if the number of species 

 of the fossil flora is assumed to be 4000, — there are actually 3945 

 described, — Palms constitute about the eighty-fourth part of the 

 whole. In the transition rocks no remnants of Palms have as yet 

 been discovered ; in the Carboniferous formation five species have 

 been noticed ; in ' Kupfersandstein,' or the Permian formation, 

 only two species have been collected ; in the secondary class of 



