264 METAMORPHOSIS 



of fungalhyphae androothairs, studiedby HABERLANDT (1889), ZACHARIAS (1891), 

 and REINHARDT (1899). So long as these structures are allowed to grow. without 

 any interference, no deposition of new lamellae and bursting of older layers can 

 be observed in them. Such a negative result taken by itself certainly does not help 

 us much, but we may obtain an explanation of the mode of growth, such as has 

 been demonstrated in Bornetia and Derbesia, if the cessation of growth be studied. 

 Thus ZACHARIAS has shown that the roothairs of Chara, when placed in water 

 in which Chara has never been grown, at once cease to grow. When, after 

 a certain interval, growth begins again, one sees that the parts of the cell-wall 

 which previously had shown power of growth, have now lost that power and 

 are burst asunder by newly-deposited layers. REINHARDT observed the same 

 phenomenon in roothairs which had been plasmolysed. Although REINHARDT 

 perhaps goes too far in assuming that in Bornetia the bursting of a layer always 

 follows a previous interruption in growth, one is forced to admit that in the 

 cases studied by REINHARDT and ZACHARIAS, growth may be normally carried 

 on without successive deposition and bursting of lamellae. The burstings must 

 take place in these examples with quite remarkable frequency ; according to 

 REINHARDT at least once or twice a minute in the case of fungus cells. It is 



scarcely possible that the mode of growth 

 can in this case be other than by intussus- 

 ception, i.e. by the interpolation of new 

 wall substance between the older mole- 

 cules. An observation of ZACHARIAS, fol- 

 lowing directly on NOLL'S experiments, 

 supports this view. He succeeded in intro- 

 ducing congo-red, under certain condi- 

 tions, into the membrane of the roothair 

 without occasioning an interruption in 

 growth, and was able to determine clearly 



Fig. 54. Development of the megaspore of Sela- that the Coloured apCX gradually lost its 



V1* M? e JgffiAS colour duri "g surface growth while older 



stages, similarly magnified (x 180). ex, exosporium ; portions of the Same Wall retained it. It 



K^SS?^^ 1 SS^Si^S!!^ m was very apparent that in this case there 



was no bursting of the coloured lamellae 



as in NOLL'S experiments, but rather an interpolation of materials between 

 the minute older particles. This result admits of more than one interpretation, 

 and hence we cannot conclude with certainty that the mode of growth in 

 Caulerpa in NOLL'S experiments is conditioned only by the colouration with 

 prussian-blue and by artificial interruption of growth. 



Further important evidence in favour of surface growth by intussusception 

 has been brought forward by ASKENASY (1890) and STRASBURGER (1889). It 

 is impossible to quote all these proofs at present, and we will, therefore, note 

 only a special case which has been recently drawn attention to by FITTING 

 (1900). The young spore of Selaginella possesses two chemically different mem- 

 branes, the exosporium and the mesosporium (Fig. 54, ex and mes) ; they are 

 separated from each other by a fluid containing very little solid material. As 

 the spore increases in size (Fig. 54, 1-III) these lamellae retain their relative 

 positions but increase correspondingly in surface growth and in thickness. In 

 an ordinary cell-wall the innermost lamella at all events can increase in surface 

 as a result of plastic stretching and simultaneous deposition of new layers from 

 the protoplasm ; the outer layers, however, must become thinner as a result 

 of plastic extension. In the spores of Selaginella, however, the inner lamellae are 

 quite debarred from any possible deposition of new layers, since the protoplasm 

 is contracted into a sphere within the mesosporium (Fig. 54, //), and touches 

 the membrane at most at one spot only. Between the protoplasm and the 



