35 



and individual flower so that there are successive stages in 

 the development of complete peloria.'^ For example : one flower 

 (Figures 2a-c) was normal except for small spurs on the lateral 

 corolla segments of the lower lip, the fifth stamen being repre- 

 sented by a minute tab of sterile tissue attached to the base of the 

 corolla. Another (Figures 3a-c) was essentially the same, except 

 that the fifth stamen, although smaller than any of the others, bore 

 two pollen sacs. In complete peloria (Figures 4a-e) five spurs and 

 five equally developed stamens are present. 



As has been mentioned in the brief description of the normal 

 flower, the single spur is found on the middle one of the three seg- 

 ments of the lower lip. It may also be noted in the normal flower 

 (Figure 10c) that it is the lower lip which is refolded, forming the 

 thickened labium characteristic of this portion of the corolla. Par- 

 tial peloria (as Figures 9a-b) may have a somewhat narrowly 

 tubular corolla with the lips reduced. Where peloria is fully devel- 

 oped, all five segments of the corolla bear spurs ; also the apex of 

 the corolla has the double fold {i.e., of a type characteristic of only 

 the lower lip in the normal form) present and involving all five 

 segments (see Figures 4d-e). Apparently those factors wdiich are 

 involved in the production of the middle segment of the lower lip 

 of the normal flower, with its spur and refolded labium, have become 

 active in all five segments of the corolla. It is also interesting to 

 note that this example of complete peloria had five equally developed 

 stamens and a tri-carpellate ovary and thus, by a strict interpreta- 

 tion of the generally accepted family characters, could not remain 

 in the Scrophulariaceae with the other flowers of the same inflo- 

 rescence. 



It is outside the province of this paper to more than speculate 

 on the origin and evolution of various flower forms, but abnormal 

 material such as this will undoubtedly present various clues useful 

 in the ultimate untangling of many of our phyletic concepts. In 

 Linaria, it is obvious that the fifth stamen, instead of being elimi- 

 nated by the advent of zygomorphy has only been suppressed, for 

 it takes but little disturbance of the normal ontogenetic pattern to 

 reproduce the structure in a completely normal form. This, then, 



^ For an account of the difficulties met in an attempt to fix the hereditary 

 condition of peloria in Linaria, see De Vries, The Mutation Theory, English 

 Ed. Vol. 2:201-220. 1910. 



