A. B. Stour 107 
It is quite probable that a larger number of pollinations during the 
early portion of the period of bloom, or an extension of tests over a 
longer portion of the time of blooming, would show that plants judged 
as self-incompatible were somewhat feebly self-compatible. The tests 
made show that plants may be highly self-fertile, partially self-fertile, or 
completely self-sterile from the very first date of bloom. If end-season 
self-fertility does develop my results are in error in that plants classed 
as self-incompatible may later have become somewhat self-compatible. 
Apparently in chicory the entire range of variability in the self-fertility 
of individuals is seen during the first few days of the period of bloom, 
Il. Evidence that self-incompatibility may lead to certain cases 
of embryo abortion. 
For plants of chicory that are fully -self-incompatible only mere 
rudiments of achenes develop and the entire head which is selfed becomes 
shrivelled as is shown in 1, 2, and 3 of Plate III. The numbers 4, 5, 
and 6 of the same plate show the well filled heads containing seeds with 
embryos from flower heads of the same plant cross-pollinated on the same 
day that the heads of figs. 1 and 2 were selfed, The conditions here. 
shown are typical of self-incompatibility, and are proof that (1) the sex- 
organs are potent and capable of functioning in certain relations and 
that (2) the plant is able to nourish embryos when there is compatible 
fertilization. It seems clear that embryo abortion of any sort, and 
especially that involving a condition of vegetative vigour, is not operating 
in such extreme cases of incompatibility. 
For the feebly self-compatible plants, however, there is usually a 
rather graded series’ of more or less developed but empty achenes as is 
shown in 7 and 8 of Plate III. At a of no. 7, is a group of 14 mere 
rudiments of achenes, at b are 5 achenes of good size but entirely 
empty, and at c is one good achene with an embryo; these are all from 
a single head of a plant judged as feebly self-compatible. No. 8 shows 
a graded series of 16 empty achenes and four good seeds, two of which 
were crushed to determine if embryos were present. 
No, 9 shows 11 good seed, 2 empty achenes which appear externally 
_ to be good seed, and 7 that are obviously shrivelled and empty. 
Thus in-the partially self-fertile plants there are usually present in 
those heads which have good seed some achenes which are considerably 
developed but which contain no embryos at maturity: Possibly the 
development of a few seeds has an influence on the phenospermic (using 
term of Goodspeed, 1915) development of certain others of the head 
