MR. W. GRIFFITH ON THE GENUS GNETUM. 305 



same parts in ordinary ovtila, precludes the necessity of any remark ; neither is there any 

 feature in the embryo, excepting its enormously long funiculus, that would lead me to 

 suspect any material deviation from the usual structure. 



The only genera of Coniferce with which I have been able to compare Gnetmn are 

 Agathia and Thuja, in both of which the nucleus has only one envelope, its cavity being 

 in Agathis lined with a very distinct amnios. The males of neither of these genera, how- 

 ever, exist in the Botanical Garden at Calcutta ; and neither of them appears to possess 

 that power, which exists to such a degree in Cycas and Gnetmn, of continuing the deve- 

 lopment of the ovulum independently of fecundation. With Cycadece I have had ample 

 opportunities of comparison, so far, at least, as regards the young ovula and perfect fruit. 

 The mutual resemblance is, indeed, so strong, that I have little hesitation in affirming 

 that the fruit of Cycas differs only materially from that of Gnetmn in the absence of 

 an additional envelope, and perhaps in the presence of the remains of the amnios. With 

 regard to the ovula of Cycas, the nucleus has, as Dr. Brown has stated in the Memoir 

 referred to, p. 24, only one envelope ; but I have no doubt, both from the difference in 

 its tissue, in Avhich there is, indeed, an obvious line of demarcation, as well as fi-om the 

 disposition of its vessels, that it consists of two, united throughout their whole extent. 

 The nucleus is likemse united by its lower poi"tion to the envelope, — its apex, which is 

 conical, being free, and prolonged into a membranous point, which is engaged within 

 the lower part of the canal formed by the opening that exists through the coat. This 

 membranous portion is tubular, the cavity extending some distance within the actual 

 substance of the nucleus : its walls are cellular, and not lined by any membrane. 



The body of the nucleus contains a large globular cavity, in which a sac of a corre- 

 sponding size and form exists. This sac (the amnios) is, at the period referred to, mem- 

 branous, and is rendered tense by an abundant gelatinous fluid ; it is easUy detached, and 

 appears to be merely applied to the walls of the cavity. 



I hope to be able at some future period to enter into more detail on the subject of the 

 development of the ovula of Cycas, which appears to promise several new and highly 

 interesting facts, particularly as regards the first appearance of cellular tissue within 

 the sac of the amnios, while at the same time its great size renders it peculiarly fitted for 

 observation. 



It is probable that the fruit represented by Dr. Hooker (Bot, Mag. new ser. vol. ii. 

 t. 2827) as that of Cycas circinalis, belongs to another species. The true C. circinalis 

 has, as Richard has stated, a fungous envelope suiTounding the albumen, which is of 

 considerable thickness towards the base of the fruit, or rather seed ; a thin membrane, 

 probably the remains of the amnios, being interposed, and adhering to the fungous en- 

 velope, except towards its apex. In this species the cotyledons are, as M. Bichard has 

 likewise stated, united, except at their bases, where a free exit is allowed for the plumule 

 on both sides. Dr. Hooker's description, particularly as regards the coats, agrees much 

 better with C sphcerica, Roxb, The sacs existing in the apex of the albumen, first pointed 

 out by Dr. Brown *, are in C. circinalis permanent ; that in which the embryo is developed 



• Prodr. Florae Novse Hollandite, vol. i. p, 347. 

 VOL. XXII. 2 S 



