found in the Oolite Quarries of the Isle of Portland. 397 



and are most perfect near the summit of the trunk* : but lower down^ the oldest 

 of them lie horizontally and at right angles to the trunk, which they entirely 

 encircle, and are more and more compressed as they are nearer to the root. 

 In form, position, and structure, as far as it can be recognised, these parts re- 

 semble what we find in the recent Cycadeae. As yet no leaves have been found 

 adherent to any of the fossil plants. At the summit of most of them there is a 

 cavity, the contents of which appear to have been removed before petrifaction 

 took place : the relation of this cavity to the trunk and leaves may be seen 

 by comparison with living plants of Zamia horrida, Cycas circinalis, and 

 C. revoluta, referred to in the note f . Although the trunk is so tall in Cycas 

 circinalis as sometimes to exceed thirty feet in height, it rarely attains six 

 feet in Cycas revoluta, and in other species is much shorter. The trunk of 

 many Zamiae is also very short. In the fossil specimens yet discovered, it varies 

 from five inches to a foot in height, and from eight to fifteen inches in diameter. 

 Their different forms will be best understood by referring to the accompanying- 

 Plates |. They are generally terminated downwards by a broad flat base with- 

 out any adherent roots ; but traces of the commencement of roots, having a 

 cavity at their centre, are visible in Sir G. Grey's specimen §. At the base 

 of other specimens there is only a coarse irregular reticulation, apparently 

 connected with the commencement of roots. The interior of these stems 

 presents two varieties of structure, which we find accompanied respectively 

 by a corresponding variation of external form, size and character, in the bases 

 of the leaves. These differences are sufficient, in the opinion of Mr. Brown, 

 to establish a new family, to which we may refer the two extinct species 

 before us, the one resembling a recent Zamia, the other resembling- a recent 

 Cycas ; and to which, from its near approximation to the existing family of 

 Cycadeae, the name of Cycadeoidece seems appropriate. 



In the largest and most abundant species {Cycadeoidea megalophi/lla\\) the 

 bases of the leaves vary from one to three inches in length, having nearly the 

 form and size of those of Zamia horrida, and measuring from one fourth of 

 an inch to an inch in the shorter, and from one to two inches in the longer 

 diameter of their transverse section. The trunk is short and flat, and exhibits 

 a deep cavity at its summit resembling a bird's nest. 



The origin and use of this cavity is rendered perfectly intelligible by com- 



* Plate XLVII. fig. 1. 



+ See at Plate XLVI. fig. 4. a plant of Zamia horrida; and in Curtis's Botanical Magazine 

 for June 1828, Plate 2826, a male plant of C. circinalis in flower; and in Transactions of the 

 Linnean Society, vol.vi. PI. 29. a female plant of C. revoluta in fruit. 



t Plate XLVII. XLVIII. XLIX. 



§ Plate XLVII. fig. 2. II Plate XLVII. and XLVIII. 



