REJUVENESCENCE IN NATURE. 199 



starch-granules originate by external formation of layers 

 around a nucleus, than with that set up by Miinter and 

 Nageli, of an inner (centripetal) formation of layers.* 

 Examining the behaviour of these starch-grains at the 

 commencement of propagation, we find them disappear 

 in the inverse order of the origin of their parts, the 

 envelope first dissolving, the nucleus remaining distin- 

 guishable for some time, till at length this likewise 

 vanishes without leaving a trace. This entire process of 

 solution is ordinarily completed in one night ; the forma- 

 tion of the gonidia and the above-described softening of 

 the cell-membrane follows at its heels. I have observed 

 an exactly similar disappearance of the starch-granules 

 shortly before the commencement of the formation of the 

 gonidia in Cladophora glomerata, the cells of which, as 

 in Hydrodictyon, contain a great abundance of starch- 



* Sohleiden (' Grundziige,' 3te Auft., p, 187, Transl, pp. 11, 567) ; Miinter 

 (' Uber das Amylum der Gloriosa snperba,' &c. ; ' Bot. Zeit.,' 1845, p. 198) ; 

 Nageli (' Zeitschrift,' 1847, p. 117). In spite of the many researclies upon 

 starcli we possess, the origin and development of the starch grain requires 

 a new and careful investigation, since none of the views hitherto put forth 

 are sufficiently supported by direct observations. The theory of centripetal 

 lamination is apparently borne out by the greater softness of the inuer 

 layers, but I would suggest that the cavity of the concentric starch grains 

 is always so small that, taking into account the enormous enlargement which 

 the starch grain undergoes during the formation of the layers, an expansion 

 of the outer layers after their formation must be assumed, such as is 

 scarcely conceivable. The small starch-granules which so frequently occur 

 among the layers, seem to possess no cavity at all ; in the sometimes single, 

 sometimes associated starch granules found so frequently in the interior of 

 chlorophyll vesicles, (I found 10 — 12 of them in one chlorophyll-vesicle of 

 the uppermost foliar joints of Chara hispida), no trace of a cavity can be 

 distinguished. Hence tlie conjecture may be admissible that the cavity in 

 tLe larger granules is not an original but a secondary phenomenon, resulting 

 from the disappearance of the nucleus. If the starch grains are not cells 

 themselves, but bodies secreted by the cell-contents, like the cell-membrane, 

 with which they are so closely allied chemically, thin conoentrical formations 

 from the outside is by far the more probable. I would compare their origin 

 with that of pearls in the oyster shell, disregarding of course the accidental 

 character of the latter. As many pearls, when in contact with their shell, 

 become covered up by the later lamellse of the shell, structures resembling 

 starch grains occur in Hydrodictyon, as abnormal formations, enclosed 

 between the lamellse of the ceU-membrane. I shall describe these more 

 minutely at anotlier opportunity, together with other strange structures 

 occurring in the cell-membrane of Hydrodictyon. 



