Fig. 14. 



(Fig. 1, Mangan, Miki-Noumura and Gross, Science 147, 

 1575, 1965; copyright 1965 by the Association for the 

 Advancement of Science.) 



Fig. 15. 



Now one's impression is certainly that a very 

 large fraction of these silver grains are either 

 on or next to the fibres. Does this mean that 

 the fibres are labeled? I think that it does, for 

 the following reason: either the fibres are more 

 radioactive than the region as a whole, or they 

 are not and the radioactivity is simply randomly 

 distributed. There are a number of ways to 

 test such a question, and the next two figures 

 show the way that we elected to do so. A sheet 

 of acetate overlay is placed on a print of the 

 type shown in Fig. 15. A circle, whose diameter 

 represents the average silver-grain diameter, 

 is drawn on the overlay over every grain, and 

 wherever a fibre occurs next to or under such 

 a grain, the fibre is indicated and that grain 

 scored as a hit. Figure 16 is the overlay pattern 

 for the print shown in Fib. 15. Next, the area of 

 the print is divided into a large number of 

 coordinates, say 10,000, and then using these 

 coordinates and the total number of silver 

 grains in the actual print, a number of points 

 is selected from a table of random numbers 

 equal to the number of grains. These points 

 will, of course, be randomly distributed over 

 the coordinate grid. Circles representing the 

 points selected from the random number table 

 are drawn on a new sheet of overlay, placed 

 over the print, and a fibre is scored as a hit 

 when it is adjacent to or under one of the circles. 

 The result is shown in Fig. 17. Now, it is 

 always observed that the number of hits obtained 

 with randomly-placed points is much smaller 



Fig. 16. 



than it is with the actual prints and grain 

 patterns. This rather rigorous test suggests, 

 therefore, that the microtubules are in fact 

 labeled. There is radioactivity in the interstices, 

 as one might have expected, but a large fraction 

 of the radioactivity, a larger fraction than would 

 be expected on the basis of chance alone, 



13 



