36 DEVELOPMENT OF LYMPHATIC SYSTEM, FISHES 



by the continuous lateral pharyngeal lymphatic (3) in the 

 twenty-two-day rainbow trout (fig. 33). The levels at which 

 the above-mentioned figures (figs. 27 to 33, inclusive) are taken 

 are indicated in figures 5, 6, 8 and 10. 



In view of these observations, which could be multiplied 

 indefinitely, I think there can be no doubt concerning the lym- 

 phatic character of these independent lymph vesicles and that 

 they cannot be regarded as artifacts. 



Injection experiments have shown that these independent 

 anlagen of the anterior portion of the lateral pharyngeal lymphatic 

 do not communicate with the precardinal vein opposite to the 

 points at which they lie. They have also shown that these an- 

 lagen do not communicate with the subocular lymph sac, nor 

 with the more caudally situated anlagen of the lateral pharyngeal 

 lymphatic which do communicate with the veins. It is evident 

 that the results obtained by the injection method do not in the 

 least degree support the view that these independent anlagen 

 have been derived from the veins. It is also evident that they 

 could not be observed in following the course pursued by the 

 injecta in the living embryo, nor in injected embryos prepared 

 by the Spalteholz method. Since these independent lymph 

 vesicles are not hollow outgrowths from the veins, there can there- 

 fore be but two possibilities regarding their origin: They must 

 either develop in situ in the mesenchyme independently of the 

 veins, or, as described by Fedorowicz ('13) and more recently by 

 Kampmeier ('15) for the lymphatics of amphibia, be formed as 

 the result of a secondary vacuolization within a solid cord of 

 cells which has grown out from some preexisting endothelium. 

 In either case the contention so often made by Huntington and 

 the writer would still be maintained, that the lymphatics are 

 formed by a concrescence of independent and discontinuous 

 vascular spaces which cannot be injected from the veins. Proof 

 that these independent anlagen are established in relation to some 

 preexisting endothelium lies in the determination of the presence 

 of an actual continuity between them and some definite endothe- 

 lial structure. Such a continuity I have thus far been unable to 

 find, either with the endothelium of the subocular lymph sacs, 



