IIII. I XDULATIXf- MI.MHRAM . 



241 



needle being about one micron in diameter at about five micra 

 back from the tip) was carefully brought up to the free margin of 

 the undulating membrane. By this means the membrane could 

 onsiderably deflected without being torn. The undulations 

 continued in -pite of the forced position of the membrane. 



With the needle it was found possible to press down upon the 

 membrane and to pierce it, Fig. 2, a. The membrane at once 

 i-plii \crtically along a line running through the puncture. Fig. 2, 



/*. Tin p. ii i of the membrane beyond the needle then rapidly 

 br.ke into a linear series of very fine cilia which \\ere plainly 

 \i-ible In -cause they were, now beating out of uni-on 1 -'ig. 2, C, 

 A- 1< >iu .1- tin- needle was kept in position t he cilia of t In- di-rup 

 pi'itioii of the membrane repeatedly fused and separated again. 

 The fu-ion was partial, being limited to several group- of the 

 cilia \\hich repeatedly kept fusing and then breaking apart. 



When the needle on the membrane was removed ihe entire n-\\ 

 ot cilia (jiiit kl\ fused and reconstituted the membrane. By !-.- 

 delicate treatment with the needle the entire membrane could be 

 bioken up momentarily into cilia. I'pon remo\al of tin- nt-rdle 

 the cilia al\\a\- reunited to form an intact, hi>mom-iu-ou- nu-in- 

 br.nn . A membrane could thus be repeatedly di -nip ted and then 



alloued to re-unu- its original normal aspect \\ithoiit any >i^n of 

 ha\ in^ been injured. This is indicated by the fact that when the 

 animal i- freed b\ removing the transfixing needle it \\ill -wim 

 aua\- with the membrane undulating normally. 



A u-mpoi-.iry. partial breakdown of the undulating membrane 

 was also ob>er\ed to incur spontaneously while under dark-field 



illumination. The picture was most striking for the membrane, 

 16 



