568 SUMMARY OF CURRENT RESEARCHES RELATING TO 



The rhizome of Dianthera americana bears many secoudary roots, 

 showing adaptation to aquatic conditions. There is a three-layered 

 esodermis, and a thin-walled cortical parenchyma of many layers. The 

 stele shows a thin-walled endodermis, a continuous pericambium, and 

 very distinct cambium arches inside the leptome. Cystoliths, sclerotic 

 cells, and crystals are absent, but a few raphidines were found in the 

 leptome of some of the roots. The polystelic structure of the stem is 

 somewhat obscured in the nodes owing to anastomosing of some of the 

 steles. In the rhizome the epidermis overlies a continuous zone of very 

 thick-walled collenchyma, followed by the cortical parenchyma containing 

 starch and raphids, but no chlorophyll. Its horizonal internodes have 

 six peripheral steles and one central, each being surrounded by thin- 

 walled endodermis containing starch. Inside the endodermis are a few 

 thick-walled stereome cells, but they do not form a sheath. In the 

 peripheral steles the leptome and hadrome form an almost complete ring 

 with a small central pith, whereas the mestome of the central stele is 

 arranged in two arches with a broad parenchyma in the middle. 

 Eaphidines are abundant in the leptome. Similar structure, with slight 

 variations in correspondence with the change in environment, is found 

 in the aerial stem. In the nodes of the latter, four of the peripheral 

 steles (two on each side) unite to form two large triangular steles, from 

 which the two leaves arising from the node receive each three mestome 

 cylinders, whereof the central may sometimes be absent. Strangely 

 enough, the axillary shoots which are inflorescences are monostelic in 

 structure. Strong evidence is adduced tending to show that the roots 

 studied by Greenish as those of Phlox Carolina, really belonged to 

 Dianthera americana. This disposes of the statement that cystoliths 

 occur in the roots of the former. 



In view of Van Tieghem and Douhot's proposal to resuscitate the 

 genus Auricula on account of its polystelic axis, the author hints that a 

 similar step may possibly be taken hereafter in respect of the polystelic 

 species (three in number) of Dianthera. It would be interesting to 

 note whether the pollen partakes in these structm-al differences. 



Reproductive. 



Embryology of Rhytidophyllum.*— M. T. Cook has made a com- 

 parative study of E. crenulatum D.C. and R. tomentosum Mart., and 

 finds that, while the two species have distinct, external differences, the 

 morphology of the embryo-sac and embryo is the same. The arche- 

 sporium is a subepidermal cell which develops directly into the func- 

 tional megaspore. There is nothing unusual in the formation of the 

 embryo-sac, and the embryo is of the Capsella Bursa-pastoris type, with 

 slight variations in the dermatogen-formation and in the basal region. 

 The endosperm is cellular, and formed in the usual way, but is soon 

 disorganised. Both endosperm and nucellus feed the embryo, and finally 

 the latter is only surrounded by endosperm. 



* Bull. Torrey Bot. Club, xxxiv. (1907) pp. 179-84 (1 pi.). 



