154 



SCIENCE 



[N. S. Vol. XLIV. No. 1127 



afford a striking contrast. The amount of 

 the plexus which can be demonstrated by 

 reconstruction increases very much if an 

 oil immersion lens is used, but the method, 

 though one of the most important aids in 

 embryology, is entirely inadequate to test 

 the method of growth of capillaries. No 

 one would regard it as adequate to deter- 

 mine an entire plexus of blood capillaries 

 even where their pattern is well known. 



It is, I think, obvious that the only ade- 

 quate method for the study of the growth 

 of capillaries is to observe them in a living 

 specimen ; and in this connection we have a 

 long series of valuable observations on the 

 classical object, the living tadpole's tail. 

 Capillaries were first seen in the tadpole's 

 tail by Schwann, and were first differen- 

 tiated into two types, blood-capillaries and 

 lymphatic-capillaries, by Kolliker. Dur- 

 ing a long series of studies with this object, 

 by Rernak, Sigmund Meyer and others, and 

 finally by Eliot R. Clark, 19 with greatly 

 improved methods, two facts have become 

 established — first, that endothelium is con- 

 tractile and second that the vessels grow 

 by the cell division of their own walls. 

 Clark was able to watch a given lymphatic 

 for several days and to observe that the 

 wall put forth tiny processes of protoplasm, 

 which we term sprouts, that the nuclei of 

 the cell divided and wandered into the new 

 sprouts, which developed into new vessels. 

 He was able to plot out every mesenchymal 

 cell in the neighborhood and to show that 

 the growing sprouts of endothelium avoided 

 rather than approached the processes of 

 mesenchyme, and never incorporated them 

 into their walls. Thus in the one place 

 where natural conditions are such that 

 every cell, or rather every nuclear area of 

 a growing vessel, and every mesenchymal 



ia Clark, E. E., Anat. Record, Vol. 3, 1909. 

 Amer. Jour, of Anat., Vol. 13, 1912. ' ' Proo. Amer. 

 Asso. of Anat.," Anat. Record, Vol. 8, 1914. 



cell can be identified, it is without question 

 true that both blood-capillaries and lym- 

 phatic capillaries grow through the pro- 

 liferation of their own walls. 



The method of growth of capillaries may 

 thus be regarded as established. But this 

 is not the whole problem for the embryolo- 

 gist. Under development he must consider 

 both the original differentiation of tissues 

 and their method of growth. In embryol- 

 ogy it has become clear that there is a 

 gradual differentiation of tissues from a 

 common cell mass, and that after a tissue 

 is once differentiated it increases by cell- 

 division. This conception of the differen- 

 tiation of tissues was clearly stated by von 

 Baer in 1828. He called the process histo- 

 logical differentiation. Thus, development 

 consists in the differentiation of tissues 

 followed by growth. The most recent work 

 on the lymphatic system demonstrates that 

 the period of differentiation of endothelium 

 is the period of the origin of the blood- 

 vessels, and that this period has long since 

 passed when lymphatics begin. Lymphat- 

 ics do not differentiate from mesenchyme, 

 but grow from veins. 



It is well known that methods have long 

 been sought by histologists to distinguish 

 endothelium from mesenchyme. If we 

 could always distinguish endothelium in 

 sections the problem would be practically 

 solved, but the difficulty of determining 

 lymphatic endothelium in the sinuses of 

 lymph glands, or vascular endothelium in 

 the spleen pulp are too well known to need 

 emphasis. These very difficulties lead us 

 to the question, is endothelium differen- 

 tiated from mesenchyme? 



Efforts to distinguish endothelium from 

 mesenchyme have not been entirely without 

 results. For example, Clark has found that 

 in the chick the nuclei of lymphatic endo- 

 thelium can be distinguished from the nu- 

 clei of the mesenchyme by characteristic 



