516 SUMMARY OF CUREENT RESEARCHES RELATING TO 



threads ; later condensation results in the formation of two rods in the 

 place of one. Reduction in number of chromosomes occurs in the pro- 

 phase of the first maturation division when members of each pair conju- 

 gate ; reduction in chromatin volume occurs in the interval between the 

 maturation mitoses, for no true resting stage is seen, and the loss caused 

 by the preceding mitosis is therefore not compensated. One maturation 

 division is reductional in Weismann's sense. 



AVhenever chromosomes have been measured the diameter of the 

 component rods has been found to be • 83 ft in phyla, including and 

 above ISTematohelminthes, 0*42;a in phyla below the latter, and 0'21/x 

 in Protozoa. These thread-widths are the same in all mitoses, for rod- 

 dimensions are constants. The rod-lengths of chromosomes throughout 

 the animal kingdom constitute the members of a series in arithmetical 

 progression, of which the difference between terms does not exceed half 

 of the greatest thread-width, i.e. 0-42/x. Complexes of closely allied 

 organisms usually show chromosomes of similar dimensions, but widely 

 separated organisms may possess certain in common, whereas sister 

 families may be entirely unlike in this respect. No correlation exists 

 between rod-lengtbs or number of chromosomes and our classification. 



The heterotropic chromosome, when present, possesses a diameter 

 greater than that of the ordinary chromosomes, and this diameter is not 

 always constant throughout its length. It cannot, therefore, belong to 

 the general series. Throughout the growth-period it remains as a com- 

 pact and darkly staining body apposed to the nuclear membrane, and 

 passes entire to one daughter-cell at the first or second maturation- 

 division. 



In organisms whose complexes are composed of chromosome rods of 

 various lengths, species appear to be distinguishable in this respect, for 

 no two have been found in which complexes are identical ; the absence 

 or presence of a certain rod-length may possibly enable us in future to 

 determine whether an organism belongs to a distinct species or merely 

 represents a variety. 



The chromosome rods of Vertebrata are usually long, those of 

 Mollusca medium or short, and those of Artliropoda medium, short or 

 spherical ; in Annelida and Nematoljelminthes, with a few exceptions, 

 the rods are spherical. In Nemertines, the highest phylum possessing 

 the smaller chromatin thread-width, they appear to be longer than those 

 of Echinoderms, which in turn have longer chromosomes than Coelen- 

 terates. The lowest Protozoa show spherical granules, whereas normal 

 chromosomes of various lengths are observed in others of greater com- 

 plexity. The ^'olume of chromatin remains constant in all metaphases 

 of the species, with the exception of the secondary spermatocyte or oocyte 

 mitosis, in which it is halved ; the volume of individual chromosomes 

 likewise remains constant, except in the first maturation-division, where 

 members of pairs are observed to have conjugated. There seems, how- 

 ever, a general tendency for the total volume to increase as we ascend 

 the animal kingdom, and consideration of each subdivision possessing a 

 definite chromatin thread-width shows an increase in rod-lengths in the 

 higher forms ; on the other hand, the lowest forms in each sub-division 

 appear to possess complexes composed of chromosomes whose rods are 



