VIII. EFFECTS OF DEFICIENCY IN ANIMALS 303 



struck l)y the deficiency of antiscorbutics. . . . All these cells seem in the 

 absence or deficiency of antiscorbutic to be subject to atrophy and yield a 

 product which is quantitatively as well as qualitatively inferior. ... If anti- 

 scorbutic is aUogetluM- lackinj^ in the food, these cells go on living for a cer- 

 tain time and yield during this time products which deteriorate day by day 

 till the activity is completely arrested and the death of the cell ensues." 



A recent study of the morphogenesis of collagen fibers as observed in 

 cultures of chick embryo skin by Porter'' has furnished convincing evidence 

 that the collagen fi])ers do form at the surfaces of cells. He found that fiber 

 formation does not proceed at random in all directions as in the formation 

 of fiijrin in clot formation or as might be expected if collagen fibers polym- 

 erized out of an intercellular matrix. It appeared rather that "the fiber 

 mat had been organized by cells, which, in migration, had deposited new 

 fiber arrays above or below those already formed." In later studies, using 

 a culture medium (Locke bouillon) without fibrin or serum components, he 

 followed the stages in fiber production and observed that most of the ecto- 

 plasmic portion of the cell appeared to be involved in fiber production. He 

 noted that "the collagen differentiates out of the surface, through the cell, 

 and then will pull off the cell as a bundle of collagen fibrils, luiit fibers." 

 After leaving the cell as small striated fibrils 200 A. to 300 A. in diameter, 

 they appeared to increase in diameter to 500 A. or 600 A. Porter considers 

 it reasonable to assume that some of the collagen from the surface of the 

 cell may go into solution and may later condense on the fine fibrils. Thus, 

 the fundamental periodicity of the fiber is laid dow^n bj^ the cell, and with 

 the pattern thus established additional units can be laid down outside the 

 cell. Although in the Locke medium he did not obtain mature collagen, 

 there was evidence that every third of the striae would expand to give the 

 640 A. periodicity characteristic of collagen. 



(2) Directly on the Extracellular Material. According to this theory the 

 fibroblasts would be able to form the supporting substances such as collagen 

 but some colloid cement substance would be lacking. This hypothesis, ad- 

 vanced by Aschoff and Koch,'" has been strongly supported experimentally 

 by Wolbach and Howe^ and Mazou^,'*'' and with somewhat less conviction 

 by Dalldorf^'' and Hunt.^' In describing the cementing together of the fibrils 

 as similar to the setting of gel, Dalldorf states that "it is precisely this 

 phase of the formation of intercellular materials which may be completely 

 controlled by vitamin C. Thus in guinea pigs which have been depleted of 

 vitamin C, the ground substance and fibrils are present as in health but 

 fibrils of collagen are not formed. When the deficiency is satisfied, trans- 

 lucent bundles or masses of collagenous materials reappear within 18 hours." 

 Supporters of this theory hold that this period of time is too short for an 

 appreciable amount of formation of completely new intercellular sub- 



