RAINEY, ON STARCH GRANULES. 3 



that they are as much starch as the larger globules, and 

 differing from them in nothing but size. These sphe- 

 rules may either be free in their starch-cells, or conglo- 

 merated and joined together in pairs or threes, producing 

 dumb-bell or somewhat triangular forms. Sometimes they 

 are found with shreds of membrane, and at others are 

 invested more or less by an utricle. In the starch-cells more 

 remote, the granules are larger and fewer, so that their 

 increase in size is attended with a diminution in number, 

 showing most clearly that the largest are the product of the 

 union of those of an inferior size. Indeed, the number of 

 granules of a small size is such in some of the starch cells 

 that it would be impossible that they all could become 

 developed into large granules without the spaces containing 

 them undergoing a most inordinate increase in size, which is 

 not the fact ; the spaces in which the middle-sized granules 

 are lodged being about the same size as those containing the 

 largest granules. But the chief evidence in support of this con- 

 clusion must be obtained from the microscopic examination 

 of all the various forms of starch, beginning with that which 

 is merely granular, and going up to that which is most per- 

 fect. Such an examination will show that there are exactly 

 the same class of appearances to be found in starch, indicative 

 of coalescence of its particles, as are presented by the several 

 forms of carbonate of lime, whether prepared artificially or 

 occurring in organized tissues. 



Plate I contains representations of different forms of 

 starch j fig. 1 is the ordinary form of the larger granules. 

 This was taken from the immature fruit of the potato. 

 Nothing that I have examined shows the laminated character 

 of starch granules so well as these potato apples, as they are 

 called. Figs. 2 and 3, drawn from specimens of common 

 potato starch, are similar to those pointed out by Mr. E. J. 

 Quekett as the result of cell multiplication by division, a 

 view still, I believe, generally entertained by botanists. This 

 hypothesis is considered by physiologists to apply only to 

 even numbers, but fig. 4, and also fig. 5, which latter is copied 

 from Criiger's plate in the ' Journ. of Micros. Science/ for 

 April, 1854, show three granules similarly united, all as 

 nearly as possible of the same size. Now the question is 

 whether this hypothesis extends also to uneven numbers, or 

 whether these specimens are merely three granules joined 

 together, and in an early stage of coalescence. Examples of 

 this form are not uncommon. In the specimen of starch 

 from which these were taken there was no difficulty in finding 

 them, being almost as common as the pairs. This starch 



