H. BOTANY AND HORTICULTURE. 307 



cies of water-lily, and the grass by the side carpeted with the 

 selaginelle, which forms such a favorite bed for ferns in green- 

 houses. There is a specimen of rhododendron 180 feet in cir- 

 cumference, and the camellias are every where loaded with 

 fruit. The dragon's-blood-tree (Dracaena draco) grow T s well 

 out of doors, as also does the Australian gum-tree [Eucalyp- 

 tus globulus). There is a magnificent specimen of the cam- 

 phor-tree (Laurus camphora), and several of the rare Ben- 

 thamia. Several marmosets are allowed free liberty in the 

 grounds, climbing to the tops of the highest trees, and al- 

 ways returning to the house at night. At a spot on Fal- 

 mouth Harbor called Flushing, the temperature through the 

 year is even still more equable, and the establishment of a 

 tropical garden there would probably be attended with the 

 most successful results. 



SORBY OX TIXTS OF FOLIAGE. 



We have already referred to the investigations of Sorby in 

 regard to the various tints of foliage, and especially to the 

 change of color in the leaves in autumn ; and in a late num- 

 ber of Nature we find a resume by him, giving the present 

 state of his inquiries on the subject. He separates the differ- 

 ent coloring matters into five groups : first, the chlorophyl 

 group, characterized by being insoluble in water, but soluble 

 in alcohol and in bisulphide of carbon, and embracing three 

 or four species ; second, the xanthophyl group, containing 

 several species, only two of which are common in leaves, one 

 being more and the other less orange. They are character- 

 ized by being insoluble in w r ater, and soluble in alcohol and 

 in bisulphide of carbon, differing, however, from the members 

 of the first-mentioned group in having peculiar spectra ; third, 

 the erythrophyl group, comprising a number of colors soluble 

 in water, in alcohol, and in ether, but insoluble in bisulphide 

 of carbon. Those met with in leaves are more or less pur- 

 ple, are made bluer by alkalies and redder by acids; and 

 thus sometimes plants containing the same kind may vary 

 more in tint, owing to a variation in the amount of free acid, 

 than others colored by entirely different kinds. Among the 

 species some have very interesting botanical relationships, be- 

 ing so far found only in particular classes of plants. Fourth, 

 the chrysotannic group, containing a considerable number of 



