142 Mr. J. Miers on the Styracese, 



that there is never any direct connexion of the placenta with the 

 style. Prof. A. DeCandolle fully admitted this affinity with the 

 Olacacea, so much so that at one time he proposed to introduce 

 the genus Liriosma into the Stijracece^. In my memoir upon 

 the latter genusf I pointed out many circumstances in proof of 

 this affinity. I showed that, in a manner analogous to that of 

 Stromhosia, mentioned in the preceding page, the ovary of Li- 

 riosma is at first superioi', but that, by the expansion of the 

 nectarial disk that supports it, the growth is wholly downwards, so 

 that the fruit becomes completely inferior, as in Halesia, crowned 

 by its small unchanged calyx. I noticed also the occurrence of 

 an epigjTious gland-like thickening of the ovary, similar to that 

 existing in HyuscyamusX, which forms a prominent feature in 

 Liriosma and Strombosia and some other genera of the Olaca- 

 cece, and occurs sometimes in Styrax and generally in Stri- 

 yilia. Dr. Asa Gray does not admit the existence of this epi- 

 gynous thickening in Styracece, saying, " It is only the ordinary 

 epidermis of the ovary, with its downy covering, unaffected by 

 the pressure of the base of the corolla and the staminal tube, 

 which closely encircles the lower part, and it readily separates 

 from the rest of the parietes, as it does also in S. Benzoin." I will 

 not affirm that it is a distinct formation ; but if it be the ordinary 

 epidermis, it certainly assumes a very thickened appearance in 

 Strigilia, projecting over the ovary like the eaves of a circular 

 roof, while the lower moiety of the wall of the ovary is attenu- 

 ated in substance §. The admission that " it readily separates 

 from the rest of the parietes," seems to favour the idea that 

 it is something more than epidermis, especially when analogy 

 points to the remarkably thick epigynous gland that ordinarily 

 surmounts the ovary in Olacacece, even when it is entirely supe- 

 rior. The concm'rence of so many points of structure in these 

 several families ought to have their due weight in the question 

 of their relative affinities. Prof. DeCandolle 1|, although he ad- 

 mitted the distant relationship of the Styracecje with the Meli- 

 aceae, yet considered that his tribe Pamphilia was more intimately 

 allied to the latter family than to the former^. Yet the same 

 features that established this identity in his opinion, exist equally 



* Prodi-, viii. 243. 



t Huj. op. 2n(l ser. viii. 103, 163; Contrib. Bot. i. 16, 23. 



+ 111. S. Amer. PI. i. 174; ii. App. i). pi. 79. figs. 8, 9-13. 



§ In Styrax officinale I do not find the same disk-like summit of the 

 ovary that I have depicted in Strigilia Icevis ; but I notice (as generally 

 throughout the Styracinea) that the lower portion of its wall, which encir- 

 cles the three-cellular portion of its base, is greatly thinner in substance 

 than its unilocular upper portion, where it is comparatively fleshy. 



II Prodr. viii. 215. 



H lb. p. 270. 



i 



