114 The Philippine Journal of Science i9i6 



the cephalic portion of the other is somewhat broken down. 

 These embryos are in the twelve-somite stage of development. 

 The medullary tubes are completely closed in the region of the 

 head and thorax, while in the caudal region they are still open 

 posterior to the sixth somite. 



The embryo on one side shows a marked disintegration along 

 the cephalic end. The caudal end is in a better state of preser- 

 vation; here the medullary groove is still open. No somites 

 appear along the lateral sides of the medullary groove. The 

 embryos are separated from each other by a cleft, which is 

 clearly seen on the cephalic end of one embryo (Plate III, fig. 2). 



DISCUSSION 



The underlying causes involved in the formation of twins and 

 other polyembryos are still somewhat obscure ; however, with the 

 experimental work that has been done in the production of poly- 

 embryony and the large amount of literature that has accumu- 

 lated on the subject, we are now able to arrive at some fairly 

 definite conclusions. It has been definitely shown that in some of 

 the anamniotes twins, triplets, quadruplets, etc., can be produced 

 by mechanical and chemical changes. These are definite results 

 that have been obtained by altering the environment by external 

 agents. If the intrinsic factors be considered, little imagination 

 is needed to conceive of distorted embryos, twins, etc., as being 

 due to faulty metabolism or to some physical law that is so wrapt 

 up in physicochemical processes that it is still impossible to 

 solve the mystery. 



Investigators have attempted to produce chick and duck twins 

 and certain types of monsters by shaking the eggs, raising and 

 lowering the temperature, or varnishing the eggshells. Results, 

 however, have been uncertain, and at no time could the experi- 

 menter foretell what kind of twins or monstrosity would develop 

 after treating the egg by one of the above methods. 



For convenience, twins should be divided into three main 

 classes — (a) dissimilar twins, (6) identical twins, and (c) 

 joined twins. The first class includes twins of different sex 

 or unlike in their general development and physiognomy. This 

 type of twins results from the fertilization of several ova or 

 possibly a multinucleated ovum. The second class includes those 

 embryos that are identical in almost every detail — the result, 

 undoubtedly, of an early cleavage of the blastomeres or a 

 complete separation of the fertilized egg at some later stage, 

 as Newman and Patterson have shown in the nine-banded arma- 



