SYSTEM AND SUPRARENAL BODIITS IN THE SPARROW. 751 



consolidation by the dispersed cells gathering into compact masses, 

 and the simultii.neousgrowth of longitudinal cellular strands, Avhich 

 unite tliem into contiuuous moniliforni chains. Silver prepara- 

 tions show scattered fibres among these cells. A graphic recon- 

 struction of the consolidated cord in the anterior thoracic legion 

 at this stage is shown in text-iig. 9 A. 



So far as we have been able to observe, the cells constituting the 

 chain at this stage do nob show any diderentiation of chromatin 

 cells among them. Their nuclei are oval, their cytoplasm darkly 

 stained and full of mii\ute granules, coll boundaiies are faint, or 

 may still be absent. We liave constantly observed the presence 

 of clear spaces surrounding the groups of sympathetic cells near 

 the dorso-lateral angle of the aorta, ICuntz has noticed these 

 clear spaces. They may be spaces filled with lymph during life, 

 or they may be due to shrinkage phenomena. We have followed 

 the development of the cords by means of graphic reconstruc- 

 tions from a series of Sparrows, separated by very slight difl'ei'- 

 ences of development, and the i^esult is striking. The series of 

 reconstructions is shown in text-fig. 9. 



We have chosen for purposes of reconstruction the anterior 

 thoracic region of the embryo, where its development is most 

 advanced. Text-figs. 9 A tt B are reconstructions of the primary 

 .sympathetic nervous system of Sparrows 7 and 2 (= 78-hr. 

 chick). They show that at this stage the sympathetic chains in 

 the anterior thoracic region consist of two approximately parallel 

 cords, formed by masses of cells united by their connecting strands. 

 Anteriorly the strands are much thickei-, but not sufficiently so 

 to hide the metamerism, Avhich corresponds exactly with that 

 of the dorsal root ganglia. In both the text-figures, but more 

 markedly in text-fig. 9 B, it will be seen that the swollen parts of 

 the cord send out lateral processes. These processes do not run 

 in the plane of tlie parallel cords, but aro directed dorso-laterally. 

 The apparent dili'erences in the directions to which they point 

 are due to the curvature of the embryo. 



Text-fig, 9 C is from Sparrow 6 (= 84-hr. chick). The monili- 

 form character of the strands is more mniked and gives the 

 impression that the swellings have increased at the expense of 

 the connectives. The distances between the ganglionic swellings, 

 e. ff. in text-figs. 9 A & 9 0, are roughly equal, but the connectives 

 in the latter have become more slender, a result which could be 

 attained either b}'^ a process of degeneration, which seems im- 

 probable at this early stage, and of which no evidence was 

 obtained in the sections, or by migration of their elements into 

 the ganglionic swellings. 



Similar migration seems to have taken place into the lateial 

 processes, which are now corinectcd by broad bases to the primary 

 ganglia, and taper oflT to a point dorso-laterally to the cords. 

 This certainly suggests that their growth is outwai'd from these 

 ganglia, a view supported by the freqtient occurrence of mitotiiig 



