552 SCIENCE PROGRESS 



of Gymnosperms was 3*S3ib*o7 mm., for the fibre-tracheids and 

 vessel segments of Dicotyledons the values were o-6i — 1-20 

 i 0-02 mm. In a recent paper Bailey {Amer. Jour. Bot.) gives 

 data respecting the comparative lengths of the cambial initials 

 and the conducting elements formed from them. The results 

 appear to indicate that the size of the tracheids in Ginkgo and 

 the Coniferae closely resemble that of the cambial cells from 

 which they are formed, the adult elements being slightly longer. 

 In Dicotyledons, however, the fibre tracheids are considerably 

 longer, though the vessel segments approximate closely in 

 length to that of the cambial elements. The ultimate size of 

 tracheary cells thus appears to be determined by the size of 

 the cambial initials as well as by the conditions of their differen- 

 tiation. 



From a study of the somatic chromosomes in Tradescantia, 

 and from that of his earlier work on Vicia {Bot. Gaz., 191 3), 

 Sharp has come to the conclusion that in these two genera, at 

 least, the chromosome split takes place during the prophase, 

 and not during the telophase {Amer. Jour. Bot.). Moreover, 

 he finds no evidence to support the contention that the 

 longitudinal split of the chromosome as a whole is in any way 

 due to the division of smaller component units. 



An extensive examination of the anatomy of the peduncle 

 in the Liliacece by Gatin {Rev. Gen. Bot.) has shown that 

 the vascular supply varies not only in different genera, but 

 is often distinctive for the different species of the same 

 genus. The vascular strands usually form alternating rings, 

 of three members each, which pass either direct into the 

 floral parts or anastomose in a ring at the base of the flower. 

 In the simplest cases there are three bundles only (e.g. Ruscus 

 aculeatus, Polygonaturn verticillatum, Scilla autumnalis). In 

 Scilla verna, Convallaria tnaialis, Allium cepa, Endyrnion 

 patiilus, etc., there are two rings of three bundles each. 

 In many cases these two sets of three bundles each can 

 be recognised, but they are followed by an outer ring, con- 

 sisting of a varying number of bundles, viz. 2 in Allium 

 ursinuni and Endyrnion nutans ; 3 in Allium moly ; 6 in 

 Agapanthus, Allium nigrum, Uropetalum, etc. ; 9 in Ornithogalum 

 umbellatum ; 1 5 in Galtania condicans, Erythronium-dens-canis. 

 In Mianthetnum, 2 bundles only are present, whilst at the other 

 extreme we have considerable complexity, as, for example, in 

 Tulipa sylvestris, where three whorls of three bundles each are 

 followed by from 36-45 small bundles. The variation within 

 the genus is well illustrated by Scilla, where the following 

 arrangements occur : S. autumnalis, 3 ; S. verna, 3 + 3 

 S. lilio-hyacinthus, 3 + 3 + 3 ; 5. obtusifolia, 3 + 3 + (2— 3) ; 

 S. ttalica, 3 + 3 + (4— 6). 



