THE DEVELOPMENT OF PATTERN lOI 



head. If the animal grows so big that part of it gets 

 out of range of the head's influence, this part may 

 develop on its own into a whole new animal, and we 

 thus get the well-known phenomenon of reproduction 

 by budding, which happens in corals, for instance. 

 Or if the head is cut out of one animal and grafted 

 into another some distance away from the host's 

 head, it can overcome the influence of the host's 

 head and start to build the competent tissues in 

 its surroundings into a complete new animal. These 

 facts can be very easily described in terms of the 

 hypothesis that there is a gradient of something 

 along the main axis of the animals, with the head to 

 the top end of the gradient and the tail at the bottom. 

 The hypothesis suggests that when any part of the 

 gradient is isolated from the rest it tends to change 

 itself back into the whole arrangement, and the head 

 or top end is particularly well able to do this because 

 it can alter any lower part which may be in its 

 neighbourhood so as to bring that part into con- 

 formity with itself. Gradients of this kind are called 

 Axial Gradients, and attention was first drawn to their 

 importance by Child. They may perhaps always 

 occur in individuation fields in some form or other. 

 But until we know rather more about what they arc 

 gradients of, they must be regarded more as a 

 descriptive convenience than as a real explanation. 



