42 FERTILIZATION 



towards regions of higher concentration of an attractive substance, 

 it is not at all clear how they do it. Does the head move to the right 

 or the left; or can the waves which travel along their tails be 

 bilaterally asymmetrical when necessary ? Or in the multiflagellate 

 spermatozoa of ferns and mosses, do some of the flagella move 

 more quickly than others under the appropriate conditions of 

 stimulation? Once the spermatozoa have arrived at the source, 

 i.e. at the tip of the pipette filled with agar and i% sodium malate, 

 they must be continually turning; otherwise they would not buzz 

 round the tip like a swarm of bees, something which every student 

 of the phenomenon has seen. The second point to notice in Fig. 

 8^ is that one spermatozoon at least, at about 2-30 o'clock (ii*i), 

 was completely unaffected by the malate gradient. 



Several workers have wrongly assumed that a substance which 

 makes spermatozoa swim more quickly, according to its concentra- 

 tion, will act as an attractive agent. The argument is that if a 

 spermatozoon happens to be swimming in the direction of ascend- 

 ing concentration of the stimulating substance, it will swim more 

 quickly and get nearer the source of the substance. If, on the other 

 hand, a spermatozoon happens to be swimming in the direction of 

 descending concentration of the stimulating substance, it will 

 swim more slowly and therefore get less far from the source. This 

 argument is fallacious, as can be seen from the following over- 

 simplified example. Suppose we have a suspension of spermatozoa, 

 moving at random (Fig. 8a), and we suddenly point a death ray at 

 a small region in the suspension. Any spermatozoon which happens 

 to swim into this region will be killed, that is to say its movements 

 will be greatly slowed up and stopped. As the movements of the 

 spermatozoa are random, most of them will sooner or later enter 

 the lethal area. In due course, therefore, nearly all the spermatozoa 

 will be found in this region, where they move most slowly. Con- 

 versely, if the region in question makes the spermatozoa swim 

 more quickly, it will on the average contain less spermatozoa than 

 there are outside. This example is over-simplified, in that the 

 lethal or stimulating region is assumed to end abruptly, and not to 

 give rise to gradients in the extra-regional space. But given that 

 this mechanism (Orthokincsis) will not achieve the desired etfect 

 and that Klinokinesis with Adaptation * is inconsistent with the 



* Orthokincsis with Adaptation can also act as an attractive mechanism, though 

 no-one has so far postulated its existence in biological systems. It suffers from 



