245 BOTANICAL GAZETTE. 



reconciled with the ])revailing theory. In the first place it is found that 

 the middle part of starch grains is first formed and the outer parts 

 deposited around it. This is proved from the fact that the corroded 

 surface of young grains is seen within older ones, covered up by sub- 

 sequent deposits. The main strength of Nageli's theory, however, 

 lies in the fact of the appearance in the starch grains of alternating 

 layers of more and less watery substance, giving the appearance of 

 concentric lines about the nucleus ; also that young grains by no 

 means resemble the center of old ones, for the latter are very watery 

 and the former very dense; also that growth in different directions is 

 unequal ; and the growth of com[)ound grains is strongest between the 

 nuclei, whereas growth by apposition would be greater at the 

 periphery. Prof. Schimper thinks that these appearances do not nec- 

 essarily prove growth by intussusception, but can be explained by 

 certain properties of starch grains. These properties are stated as 

 follows: "(i) Starch grains are rather brittle parallel to the layers, 

 but very extensible perpendicularly to them. (2) Cutting, crushing 

 or extension causes the dense starch substance to swell up consider- 

 ably and to take all the properties of the more watery parts of intact 

 grains. (3) Swelling up in water is much stronger parallel to the 

 layers than perpendicularly to them." Nageli recognizes tension, but 

 thinks it due to the intercalation of starch molecules being greater in 

 one direction than another. Prof. Schimper says the tension is due 

 simply to the unecpial swelling in different directions and that the 

 starch substance would be extended, nut broken, and hence would 

 become more watery and less refractive. Therefore wherever tlie 

 tension is greatest the less watery substance will appear more watery. 

 The tension would naturally be greatest at the central part of the grain 

 and the middle part of the layers. 



The unequal growth of starch grains is explained by the unequal 

 conveyance of material. Starch grains have central nuclei whenever 

 they are completely imbedded in chlomphyll granules. When tiiey 

 are formed at the edge of these granules they become eccentric. 

 The formation of compound grains is thought to be due to the grow- 

 ing together of free granules and not to division of simple grains, as 

 Nageli holds. 



Lililini Gl'ciyi. — I think not more than half a dozen plants of 

 this species have so far been found. Mr. Watson regards it as a good 

 species. Dr. Gray, in a letter to the writer, believes it will not be found 

 sjiecifically distinct from L. Caimdcnsc. On Roan Mountain last year 

 the writer saw a plant from which the flower had been eaten by a cow, 

 while on the stage ascending the mountain, liuton returning next day 

 could not find it again. Subsequently Prof. P:)rter found one in fruit, 

 kindly dividing it, the living root to the writer, and the top for his 

 herbarium. This root, having been dug up before mature, had but a 

 single instcnd of two flowers as last year, but afforded a chance for 

 contmuous observation. The perianth is so short and broad that the 

 first impression is that of a EritiUaria. If Liliiiiii Caiiadciisc was not 



