THE PALMS OF BRITISH INDIA AND CEYLON. 685 



Here and there may be seen a little spheroidal cavity ; finally the 

 tubes appear to be each tipped with a small swollen head. 



" Throughout the albumen the above-described structure is more 

 or less regular, ofiering a beautiful study to the Vegetable-anato- 

 mist. 



" Generally speaking, the starry cavities are arranged in a 

 quinceux, so that the interval between two of them corresponds to a 

 third. A little attention also enables the observer to see that those 

 rays, which are terminated by a little head, answer alwaj^s to one 

 another. The space between these heads is largest in dry a slice and 

 least in a moist one. The central cavity is sometimes emptj^, and 

 sometimes filled with a substance composed of very minute globules, 

 reaching occasionally to the very tips of the rays. 



" It is evident that these starry cavities represent so many hollows 

 of cells, which still preserve their radii of communication, though 

 the primitive parietes are obliterated ; and in some instances, the 

 cavities only remain in the form of ovoidal cells, which still can be 

 restored to their original configuration by immersing the portion of 

 albumen in Canada balsam. I have dried a carefully cut slice of the 

 ivory and then subjected it to the influence of Canada balsam, which 

 rendered it so perfectly diaphanous as to be scarcely discernible by 

 the naked eye from the balsam. This process has the effect of res- 

 toring the cells to their normal structure ; they become 6-rayed, 

 the tubes correspond exactly with each other and every one is tipped 

 with a swollen head, and more or less filled with the globular subs- 

 tance of which I hav^e spoken. Thus we see revealed the whole 

 organization of vegetable ivory, which is merely a prismenchyme 

 with thickened cells, in which the rays of communication are pre- 

 served. The closest scrutiny has not enabled me to detect in the 

 thick portion of the cells the smallest trace of those layers of growiih 

 which have been detected by M. Valentin, especially in the lioya 

 carnosa and Oreodoxa regia, etc. 



" This substance, which appears continuous, is very analogous to 

 that which MM. Schleiden and Theodore Vogel, in their researches 

 into the nature of the albumen, have found in the albumen 

 of PJuem'x dactylifera ; only that in the latter, there is no starry 

 disposition of the tubes, and the hollows of the cells are elongated 

 into two, or, at most, into three radii of communication." 

 (Morren). 



Cultivation in Europe. — This palm thrives in a compost of 

 two parts rich loam, one of peat, and one of river sand. Perfect 

 drainage and a liberal supply of water are essential. 



Illustration. — The two photographs reproduced on Plates XOIII 

 and XCIV were taken by Mr. Macmillan in the Botanic Gardens 

 of Peradeniya, Ceylon. 



