IRbobora 



JOURNAL OF 



THE NEW ENGLAND BOTANICAL CLUB 



Vol. 8. March, 1906. No. 87. 



HYBRIDISM IN THE GENUS VIOLA. — III. 



(Plates 66-70.) 

 Ezra Brainerd. 



It is convenient to liave some simple rule to determine the order in 

 which the parent species shall be given in naming a hybrid; on the 

 whole the alphabetical order adopted by the ( )ld World botanists has 

 most to recommend it. The eight hybrids discussed in Rhodora, vi. 

 213-223, thus amended, are as follows: — 



1. Viola fimrriatui.a X septentrionai.is. 



2. V. cucullata X fi.mbriatula. 



3. V. FIMBRIATULA X SORORIA. 



4. V. AFFINIS X SERTENTRIONALIS. 



5. V. CUCULLATA X SEI'TENTRIONALIS. 



6. V. SEPTENTRIONALIS X SORORIA. 



7. V. AFFINIS X SORORIA. • 



8. V. CUCULLATA X SORORIA. 



1 would first add to this Hst three other Vermont hybrids, — the 

 three possible combinations of V. affitiis, ]'. cucuUata and V. ne/ph- 

 rophylla. A grouj) of three thus interrelated we may speak of as a 

 triad. 



9. V. AFFINIS X CUCULLATA. — This differs from V.affinis in hav- 

 ing larger leaves, in having a dark-blue ring around the center of the 

 open corolla, in having long slender apetalous flowers, and in having 

 long-auricled sepals; in these four respects it approaches V. cucullata. 

 It differs from W cucullata in having subacuminate leaves, in having 

 purphsh petals, in having a somewhat bearded spurred petal, and in 

 having cleistogamous capsules dotted with brown; and in these four 



