1905.] NATURAL SCIENCES OF PHILADELPHIA. 183 



boundaries of each univalent component at this stage, but from condi- 

 tions in later stages (his figs. 9-22) concludes that the middle point of 

 each double rod is such a boundary. This is the weak point in his 

 whole analysis. For why may not each of the single rods be a whole 

 univalent chromosome, the bivalent One being formed then by a con- 

 jugation side to side of two univalent ones? None of his figures in the 

 least exclude this possibility. And here may be recalled Lerat's (1902) 

 somewhat inclusive observations also upon Cyclops, to the effect that 

 each such double rod may be so constituted. Then there is a great 

 break in Riickert's observations between the stages of his figs. 8 and 9, 

 the one apparently a post-synapsis or equivalent early stage, the latter 

 an advanced prophase; yet within just this undescribed period we 

 would expect great changes in the form of the chromosomes, such, for 

 instance, as the appearance of an indubitable longitudinal split. He 

 states that each double rod in the prophase bends at the middle, and 

 later breaks transversely into two at this point; this he conceives to 

 be a separation of the univalent chromosomes at the point at which 

 they had hitherto been united. But he presents no positive evidence, 

 certainly not in his drawings, that this is not the appearance of a longi- 

 tudinal split of each univalent element, a split whereby the halves 

 would remain attached at one end and gradually separate at the other 

 (just as has been described for other objects), opening up from a narrow 

 V to a condition in which the separated halves of one univalent chromo- 

 some come to lie together in one straight line. The uppermost chromo- 

 some of his fig. 116 is evidence of such a possibility. Accordingly, 

 though the bivalent chromosomes lie so in the equator of the first pole 

 spindle that their long axes coincide with this plane, and their "trans- 

 verse " splits are at right angles to it, this does not prove the first mitosis 

 to be equational. For all the proof he brings to the contrary, the open- 

 ing along the length of each bivalent chromosome may be a line sepa- 

 rating its two univalent components, and the first division therefore 

 reductional. Riickert does not convince, though his is in many re- 

 spects the most careful work yet done upon these forms, because of 

 the hiatus in his stages, and because, and this is the cardinal issue, he 

 failed to decide the mode of origin of the bivalent chromosomes. 



I have not at hand Hacker's earlier papers (1891, 1892, 1893), and 

 so quote from Riickert (I.e.) his opinions: "Zuerst hat er die Reduk- 

 tion in die erste, dann in beide und zuletzt in die zweite Teilung ver- 

 legt." In later papers (1895, 1899) he confirms Riickert's view that 

 the reduction is effected in the second maturation mitosis; though I 

 cannot see that in so doing he brings any stronger proof than did 



