DEVELOPMENT OF VESSELS. 



91 



threadlike processes will have to be regarded as peculiar structures,, 

 which serve to map out the course of the future vessels. Long 

 and patient investigations have led to the conclusion that either 

 alternative is possible. The capillary protrusions neither seek 

 nor avoid the corpuscular elements. The direction of the future- 

 vessel is determined bv conditions of a more ^-eneral order ; 

 should it happen to coincide with a neighbouring cell, this takes- 

 part in forming the wall of the vessel ; should it coincide witli 



Fig. 



Tertiary vascularisation. Border of tadpole's tail. a. Capil- 

 laries ; l>. Lymphatics ; c. Vascular buds ; d. Ditto, in 

 connexion with a connective-tissue corpuscle; c. Free 

 border, with epidermis. 



the boundary-line between two adjacent cell-territories, we see- 

 the vascular bud taking an independent course. According to- 

 Strieker J the outer surface of the capillaries is continuously coated 

 w4th a thin layer of protoplasm ; the protoplasm is not confined 

 to the nucleated points ; and this view would at once provide us- 

 Avith materials for the development of our vascular buds. That 

 these really consist of jorotoplasm is shown, on the one hand, by 

 their optical similarity to the protoplasmic processes of the con- 

 nective-tissue corpuscles; on the other, by the fact that these 

 processes can take their place in the development of new vessels. 

 When the new capillary is thrown open to the blood-current,. 



