62 



thin distinct!}^ yellow layer; i. e., there was several times as much 

 growth as in the check tube, but there was no visible diastasic action. 

 The growth of Ps. jyhaseoll on the check was now at least 100 times as 

 abundant as that of Ps. hyacinthi on the same medium. 



On the starch jelh^ with addition of the dextrin Ps. campestris and 

 Ps. phaseoli both made a good g-rowth. On the seventh day Ps. 

 (Ximpestris covered the whole surface of the long slant to a depth of 1 

 to 3 millimeters with a semifluid, smooth, wet-shining slime, and the 

 diastasic action now involved nine-tenths of the starch. The conver- 

 sion of the starch was clearly visible, proceeding slowly and uniformly 

 from the surface of the slant inward. There was a distinct line of 

 demarcation between the converted and unconverted starch. The 

 latter was bluish white, opalescent, translucent, firm, elastic, insolu- 

 ble; the former was dead white, opaque, soft, inelastic, and soluble in 

 water on gentle shaking. This part gave no color reaction whatever 

 on adding iodine. water. On washing it all out the unchanged one- 

 tenth in the bottom of the tube was seen to have preserved the shape 

 of the original slant, and on adding the iodine water it became bright 

 blue. In the corresponding tube of Ps. pliaseoli the growth at this 

 time appeared to be equally as good, but only about two-thirds of the 

 starch Avas converted. The diastasic action proceeded from the surface 

 inward in the same regular manner, the line of demarcation between 

 converted and unconverted starch was equally sharp, and the converted 

 portion had all the peculiarities recorded for that acted on by Ps. 

 camjMstris. A fragment of this soft white starch as big as two peas 

 was stirred up in 5 c. c. of the ver}^ sensitive pale brown alcoholic 

 iodine water, but no color reaction could be obtained. This changed 

 starch included all of the outer 5 or 6 millimeters of the slant; on fill- 

 ing the tube part full of water and shaking gentlj^ all of it dissolved 

 readily, leaving in the bottom a translucent, bluish white, insoluble, 

 miniature slant, which immediatelv reacted l)right blue on pouring in 

 the same iodine water. These experiments show that the presence of 

 albuminoids is not necessary for the production of the diastasic fer- 

 ment and also that it is excreted by these two species in the presence 

 of an abundance of readil}^ assimilable food. 



On the twelfth day, in the check tube of Ps. hyacinthL the thin, 

 pale yellow gj-owth had extended over most of the slant surface, but 

 it was still not one-hundredth part as abundant as in the correspond- 

 ing tube of Ps. pliaseoli., and there was no evidence of an}^ dia- 

 stasic action, whereas in the latter more than nineteen-twentfeths of 

 the starch had been digested. In the tube wdiich received the dextrin, 

 Ps. hyacinthi had made, on this date, a good, bright yellow but rather 

 dry growth over the whole surface. On the thirtieth day, in this 

 same medium, there was a plentiful, smooth, wot-shining, ])right yel- 

 low slime over the whole surface, i. e., growth enough to hide the sub- 



