480 SUMMARY OF CURRENT RESEARCHES RELATING TO 



somes of animal cells, and which appear to be an essentia] part of 

 embryonic cytoplasm. With cell-development the cbondriosomes are 

 gradually converted into simple nuclei, nuclear threads, etc., the most 

 remarkable change being their conversion into chloroplasts and leuco- 

 plasts. Pensa and others have independently made a similar discovery 

 in certain living plant-cells, but, contrary to the present writer, regard 

 the chondriosomes as of nuclear origin. 



Heterogamic gemini in Impatiens.* — J. Granier and L. Boule 

 have studied the chromosomes of Impatiens glanduligera Royle, and find 

 that, in addition to the ordinary chromosomes, the somatic cells possess 

 two macrochromosomes, while the male and female cells possess only one. 

 The fertilized ovum contains two macrochromosomes, so that one must 

 be derived from the male cell and the other from the female. The 

 authors find that by association of two double chromosomes a macro- 

 chromosomic group (gemini) is formed during the first stages of reduction- 

 division, and in this group a chromosome of male origin always unites with 

 one of female origin. Finally each sexual cell possesses one macrochromo- 

 some derived from the macrochromosomic group. 



Carbohydrates in Seeds. f — E. Schulze and M. Pfenninger have 

 made further experiments upon lupeose, but have been unsuccessful in 

 proving it to be a simple substance, although there is every probability 

 that this is the case. Numerous preparations of lupeose were made from 

 different seeds and by different methods, but they were all of the same 

 chemical nature, and the products resulting from hydrolysis were a 

 glucose, galactose, and fructose, thus proving it to be a polysaccharide, 

 probably a tetrasaccharid. All attempts to obtain crystals of it have 

 been unsuccessful, thus indicating a difference between this body and 

 stachyose, which it greatly resembles in many ways. Schulze and Godet 

 have described two carbohydrates prepared from Phaseolus vulgaris, and 

 experiments seem to show that they contain lupeose. 



Structure and Development. 



Vegetative. 



Phloem of Juglandacese.J — A. F. Hemenway publishes his first paper 

 dealing with the phloem of Dicotyledons. The present work deals with 

 six species of Juglandacese, and the author finds that the lateral sieve- 

 plates have the same appearance and structure and seem to have the 

 same function as the sieve-plates on the end-walls. The callus also 

 appears to be identical. The large sieve-tubes have as large and numer- 

 ous lateral sieve-plates as terminal ones. This is contrary to ordinary 

 views, and it seems as though this is an ancestral character indicating 

 that such lower "Dicotyledons as the Juglandaceae are nearly related to 

 the Gymnosperms or Vascular Cryptogams. Possibly further study of 

 this feature may be of use in a systematic classification from an anatomi- 

 cal standpoint. 



* Coinptes Reudus, clii. (1911) pp. 1020-22. 



t Zeitschr. Phvs. Ciiem., lxix. (1910) p. 366. See also Bot. CentralbL, cxvi. 

 (1)11) p. 475. % Bot. Gaz, li. (1911) pp. 130-5 (1 pi.). 



