214 



SCIENCE 



[N. ». Vol. XXVII. N'o. 684 



less regularly upright and parallel and 

 then proceed to cut off sterile cells at their 

 tips. The sterile tips push up against the 

 epidermal cells and soon degenerate. A 

 more or less prolonged period of vegetation 

 appears thus to intervene before the con- 

 jugations begin. (2) Generally only one 

 of the two conjugating gametes bears such 

 a sterile tip while the other shows no such 

 differentiation. (3) The gamete which 

 bears the degenerating tip cell often ap- 

 pears to be placed somewhat above the 

 other, thus suggesting that the earlier 

 hyphae fuse, not among themselves, but 

 with other hyphffi which push vip later 

 from below. Such a differentiation in time 

 of development does not involve, in the 

 writer's opinion, a moiTDhological differen- 

 tiation; the lower gamete is therefore not 

 to be regarded as a " vegetative cell. ' ' The 

 sterile cell, according to these views, is not 

 an abortive, functionless trichogyne, as pro- 

 posed by Blackman's theory, but merely a- 

 "buffer cell," a degenerate gametophytic 

 eeU, morphologically similar to the func- 

 tional gametes. This of course leaves the 

 so-called spermatia still unexplained. 

 The Relationships of the Mciditmi-cup 



Type of Bust: E. W. Olive. 



The recent work of Blackman and Christ- 

 man has solved to a great extent the prob- 

 lems which concern the origin of the diffuse 

 Cceoma type of rust. But the more com- 

 plicated, compacted cup fructification ap- 

 parently needs further explanation. This 

 is rendered evident by the recent discovery 

 in a nuinber of species of large, irregularly- 

 shaped, multinucleated cells, which appear 

 to arise from the stimulated growth which 

 follows the sexual fusion. Sometimes sev- 

 eral such multinucleated cells may be seen 

 at the base of the secidium cup. In some 

 of these instances they undoubtedly orig- 

 inate as detached buds or branches from a 

 single, central growth; in still other cases, 

 more than one fusion apparently takes 



place and more than one center of growth 

 thus results. It is quite probable that 

 these large, multinucleated cells may prove 

 to be similar to the archicarps of DeBary, 

 Massee and Richards. The basal cells 

 which form the origin of the rows of ascidi- 

 ospores arise as the ultimate branches of 

 the multinucleated cells. 



The secidium-cup type of rust is thus 

 derived from a deep-seated, more or less 

 limited, mass of cells; and the peridium 

 arises as an enclosing layer, apparently in 

 consequence of the deep-seated character 

 of the fructification, as well as from its 

 more or less centrifugal growth. The 

 fructifying organs of the diffuse Cceoma 

 type are, on the other hand, more or less 

 superficially placed, and no peridium is 

 formed in this instance. Many sexual cell 

 fusions, in most instances probably one for 

 every spore-row, occur in this kind of rust ; 

 whereas in the secidium cup but few sexual 

 fusions apparently take place. 



The simpler Cceoma type sometimes 

 shows in its course of development a phe- 

 nomenon which, in my opinion, points to 

 the conditions in which the more compli- 

 cated fficidium-cup fructification probably 

 had its origin. Following the sexual fu- 

 sion, the two nuclei thus brought together 

 begin rapidly to divide by conjugate divi- 

 sion. Sometimes nuclear division thus goes 

 on more rapidly than cell division ; so that 

 as a result several nuclei (as many as six 

 have been observed) come to lie in the one 

 large fusion cell. It is quite conceivable 

 that a still further development of such a 

 multinucleated fusion cell, coupled with the 

 partial suppression of other neighboring 

 cell fusions, especially in the case of a deep- 

 lying Cceoma, might give rise to the cup- 

 shaped type of fficidium, with its envelop- 

 ing peridium. This idea is made the basis 

 of the conclusion that the complicated 

 fficidium-cup fructification might be thus 

 directly derived from certain of the 



