286 Experimental Zoology 



A step further involves the union of different regions, and many 

 combinations of this sort have been made. It is a point of some 

 interest to find that complementary parts unite with no greater 

 facility than unlike parts, and even parts with reversed orienta- 

 tion unite as readily as do those having the same orientation. 



A series of important experiments in grafting earthworms have 

 been made by Joest. The cut surfaces are held together by 

 means of two or three hgatures through the skin. In the course 

 of a few days the contact surfaces unite with each other, and 

 the threads are sloughed off. Parts of different individuals can 

 be united as readily as parts of the same worm. Not only com- 

 plementary regions can be united, but cut surfaces of different 

 levels, and in this way short or long combinations can be made. 

 For instance, if the middle region of a worm is cut out, and the 

 end pieces grafted together, a "short" worm is produced. If 

 the union is perfect, no regeneration takes place where the pieces 

 have grown together. It also appears that the tail end of the 

 short worm does not continue to grow to make the worm 

 longer. A "long" worm may be made by inserting a middle 

 piece between the anterior and posterior halves of another 

 worm. This combination is also permanent, if the parts have 

 been perfectly united. 



In the preceding cases a posterior cut surface is united to an 

 antjprior cut surface, i.e. the pieces have the same orientation. 

 It is possible also to unite two anterior cut surfaces or two poste- 

 rior cut surfaces. For example, if two tail ends of two worms 

 are sewed together by their anterior cut surfaces, a permanent 

 union may be effected without subsequent regeneration, al- 

 though the combination has two tails and no head, and must 

 slowly starve to death. It is more difficult to unite two posterior 

 cut surfaces, not because of any inherent difficulty in the growing 

 together of the parts, but because the pieces tend to crawl away 

 from each other, and break the ligatures before union of the 

 tissues has been effected. 



The results are somewhat different if the cut surfaces are 

 brought together so that the median plane of one piece does not 



