1908] DO RET Y— CERATOZAMIA 205 



older plants by Worsdell (19), who seized the occasion to emphasize 

 the similarity to the Medullosae. 



In 1904 Matte (6) published his masterly thesis, which in the 

 third part treats of the seedlings of Dioon edule Lindl., Cycas sia- 

 mensis Miq., and Encephalartos Barteri Carruth. Of his two seedlings 

 of Dioon edule one proved abnormal, one cotyledon being partly 

 aborted. The cotyledonary bundles are mesarch near the base, but 

 the protoxylem moves quickly outward, so that the bundles are exarch 

 throughout most of their course. The axis is considered as an aggre- 

 gation of leaf bases, and the girdling is explained as being due to 

 intense intercalary growth produced under the influence of the 

 developing leaf within. The four root poles are inserted upon the 

 cotyledonary bundles, the median bundles furnishing insertion for 

 two diagonally opposite poles, and each lateral trace uniting with its 

 mate of the other cotyledon to furnish insertion for the other two 

 poles. Mucilage canals are reported in the root, for the first time in 

 a cycad. The root of Cycas siamensis is diarch; its two poles are 

 inserted upon the median bundles of the cotyledons, the lateral 

 bundles dying out in the cortex. Anomalous thickenings or cortical 

 vascular strands are conspicuous; Matte's demonstration that 

 these are not abnormal, but merely remnants of an ancestral charac- 

 ter, justifies his objection to the term "anomalous" used to describe 

 them. 



In Encephalartos the root is pentarch, and the cotyledonary 

 bundles — a single, ringlike trace from each of the three cotyledons — 

 fuse with the leaf traces at different levels before entering the central 

 cylinder. This cylinder is polystelic, as in Medullosa anglica Scott. 

 In considering this feature, Matte indorses Worsdell's view that 

 it helps to relate the Cycadales to ancient forms like Medullosa 

 rather than to monostelic forms like Heterangium as Scott suggests. 

 He regards the root of cycads as a "new organ" inserted upon the 

 lower extremity of the hypocotyl, and not merely an extension of that 

 organ. 



The main facts discovered by these investigations, and agreeing, 

 so far as they relate to Dioon edule, Cycas, and Zamia, with unpub- 

 lished studies made in this laboratory by Land, Thiessen, and others, 

 may be classified as follows: 



