STRUCTURE OF THE STEM OF THE SCREW PINE. 55 



contained in bundles of large ovoid cells. In a thin trans- 

 verse section the openings of these cells catch the eye at 

 once, both from their size and from the surrounding cells 

 being arranged round each of them in the form of a rosette. 

 It is rather difficult to see why the deposition of crystalline 

 matter should determine the enlargement of the coniaininjx 

 cells. It seems decidedly a physical as contrasted with a 

 vital process, and there appears to be no reason why one 

 cell should be the seat of it more than another. More 

 mineral matter, probably, passes into the plant in the water 

 taken up by the roots than it is able to use up in com- 

 bination Avith cellulose in the construction and thickening of 

 cell-walls. Some of this superfluity, united with organic 

 acids, is deposited, probably in the first instance accidentally, 

 amongst the contents of a not yet inert cell. Then, as the 

 process of deposition will continue after once beginning, the 

 cell is obliged to enlarge in order to supply the same space 

 as before for its other contents. 



The occurrence in the tissues of Pandanus of crystalline 

 forms of two kinds is remarkable, but not quite a solitary 

 instance. In the leaves of Buonapartca juncea Gulliver^ 

 has found a profusion of bundles of raphides, atul also of 

 " single four-sided prisms, flattened at the ends like a 

 mason's chisel." 



Although the screw pines are allied to the palms, accord- 

 ing to Mohl neither raphides or other crystals are found in 

 the plants of that family. On the other hand, in the family 

 of the IridacecR, which are not nearly related, Gulliver finds 

 prismatic crystals dispersed in the parenchyma to occur more 

 frequently than in any other. 



I am inclined to think that the presence or absence 

 of raphides or other crystalline bodies in plants depends rather 

 upon the texture of their tissues than upon their affinities. 

 Soft or succulent tissues appear to be favorable to the occur- 

 rence of raphides, though the rule is not without exception. 



As to the constitution of the prismatic crystals in Pan- 

 danus, I can make no definite statement, inasmuch as the 

 extreme minuteness of the crystals and the difficulty of iso- 

 lating them from their containing cells makes it all but 

 impossible to test them in any way. I think, however, 

 that there is little doubt that they consist of calcium oxa- 

 late. According to Sachs, this is the usual material of the 

 crystals found in the cells of plants. In the four-sided 

 prismatic form it crystallizes with six molecules of w^ater, 

 while in raphides only two molecules are present. 

 ' 'Ann. and Macr. Nat. Hist.,' v. xiii, lOS. 



