Anatomie. 371 



throughout their length , and the structurc compared with that of 

 related fossil plants. In view of the facts brought out, it is conside- 

 red that a relationship between Cycadales and Lyginodendron is not 

 sustained, but that a Medullosa-like plexus ot Paleozoic forms has 

 given rise to several lines, to one ol which the cycads belong. 



M. A. Chrysler. 



Thompson, W. P., On the Orisjin ofthe Multiseriate Ray 

 ot the Dicotyledons. (Ann. Bot. XXV. p. 1005—1014. With 2 

 plates. 1911.) 



The object of the author was to discover the origin of the mul- 

 tiseriate ray and to establish its position in the general scheme of 

 ray development in the Dicotyledons. The family Ericaceae was 

 studied in detail and confirmatory evidence was obtained from the 

 Casaarinaceae , Fagaceae and Betulaceae. 



The study of ray formation in the Ericaceae showed that multi- 

 seriate rays are commonly formed by the breaking up of Compound 

 rays. In some species, e. g. Rhododendron punctatum, only one 

 group of multiseriate rays (formed from one Compound one) can be 

 seen in a given section. The multiseriate rays may become uniformly 

 distributed throughout the wood either by several traces occurring 

 at the same level, by rays from traces at different levels overlap- 

 ping, by groups of rays becoming vertically extended as the stem 

 increases in size, or more especially by the acquired habit of for- 

 ming multiseriate rays independently of the leaf trace. 



The evidence afforded by the Ericaceae is in agreement with 

 the observations made on the other families cited, and leads to the 

 conclusion that the multiseriate ray represents the most recent 

 development in ray structure among the Dicotyledons, and that it 

 originates by the breaking up of Compound rays. Reversion to the 

 ancestral Compound type of ray may occur in the seedling, root, 

 etc., of dicotyledons which characteristically possess the multise- 

 riate ray. 



The replacement of the Compound rays by a System of smaller 

 multiseriate ones in arborescent dicotyledons has probably resulted 

 from the acquirement of the deciduous habit, the System of smaller 

 rays affording equally large storage capacity and allowing of a more 

 convenient general relation, between conducting, supporting and 

 storing tissues. E. de Fraine. 



Wille, N., Om Stammens og Bladets Bygning hos Myrio- 

 carpa cordifolia Liebm. [On the structure of stem and 

 leaf in Mxriocarpa cordifolia]. (Biol. Arb. tilegn. Eug. Warming, 

 p. 266—279. 12 fig. Köbenhavn 3 Nov. 1911.) 



Anatomical description of a mexican Urticacea cultivated in 

 Christiania. Small chloroplastids and starch-grains are found in 

 the wood-parenchyma, and in the libriform cells, the latter are 

 rather thin-walled. The vessels of the secondary wood have irregulär 

 ring-pores arranged in groups. The formation of lysigenous mucilage- 

 channels is described; the plant is very rieh in mucilage; and as 

 the plant is not at all xerophilous the author means the mucilage 

 cannot be a reservoir for water as has been suggested. There is 

 also only a small amount of water in the mucilage of the not 

 wounded plant, and this water must therefore be eagerly retained. 

 Some cystoliths in the leaf-epidermis are like those described 



