86 Papers from the Marine Biological Laboratory at Tortugas. 



vertebrates is very incomplete. Favaro (1906) and Allen (1907, 1908) 

 have done much to clarify our knowledge concerning the relations 

 between the lymphatic and the blood vascular system in fishes. Much 

 more work of this kind is greatly needed to fill up the many gaps still 

 existing in our knowledge of the lymphatics of lower vertebrates. The 

 literature of the ontogeny of the lymphatic system of lower vertebrates 

 is still more scanty. Outside of the mammals, it is limited almost entirely 

 to the paper by Sala (1900) on the lymphatics of the chick, and by 

 Knower (1908) and others on the frog. 



The injection of the lymphatics with India ink gives very beautiful 

 demonstrations of the spread of the lymphatic ducts from their point 

 of origin. But this method alone does not prove conclusively that they 

 are the continuous peripheral outgrowths of certain venous buds; since 

 if the ducts are formed by the confluence of independent mesenchymal 

 spaces it is more than likely that this process proceeds peripherally from 

 the several primary foci; hence the injections w^ould merely show the 

 several stages of progress. A few cat embryos were successfully injected 

 by this method, but have not been studied in sections. The writer hopes 

 to prepare a series of turtle embryos by this method during the present 

 summer for comparison with the reconstructions prepared from serial 

 sections. 



The choice of a suitable fixing agent is a very important factor to 

 be taken into account. In turtle embryos, where the integument is so 

 thick and the tissue composing it is so very compact, this becomes very 

 apparent. In embryos fixed with corrosive sublimate mixtures, it was 

 impossible to study the intercellular mesenchymal spaces with any 

 degree of satisfaction. 



The results of the investigation of the development of the posterior 

 lymph hearts of the loggerhead turtle by the means of serial sections 

 seem to indicate that the mesenchymal spaces play a much greater part 

 in the development of the lymphatics in general than is usually supposed. 

 It appears as though these spaces had captured, as it were, certain capil- 

 laries and had converted them into the anlagen of the lymph hearts. 

 This process takes place very rapidly, even in the turtle, so that it might 

 easily be overlooked. As soon as the mesenchyme begins to condense 

 around the veno-lymphatic spaces the process is so masked that, if it 

 still continues, it is no longer recognizable. 



SUMMARY. 



(1) The development of the posterior lymph hearts of turtles is 

 initiated by the vacuolation of the postiliac mesenchymal tissue during 

 the middle and latter part of the second week of development. 



(2) The spongy tissue thus formed is then invaded by capillaries 

 from the first two or three dorsolateral branches of the caudal portion 

 of the postcardinal veins. 



(3) Near the close of the third week, parallel veno-lymphatic chan- 

 nels are formed in this spongy area by the confluence of the mesenchymal 

 spaces with each other and with the invading capillaries. These channels 



