1 62 Blackburn and Harrison . — The Status of the British 
The homotype divisions succeeding this call for no special remark, 
so‘great are their uniformity and fidelity to type. Perhaps we should state 
that the two spindles, for the most part, as shown in PL IX, Fig. 18, lie at 
right angles to one another. 
As the chromosomes reach the poles they are grouped together, but 
immediately separate, becoming granular and once again uncountable in 
the process (PI. IX, Fig. 19). Once more, too, the nucleolus becomes an 
evident adjunct of the nucleus. Finally, and with but little delay, the 
mother-cell divides into four spores, which with the secretion of a cell 
wall bring into being a perfectly ordinary tetrad and, later, functional pollen 
(PI. IX, Fig. 20). 
Ordinarily a brief account of the course of events in the meiotic phase 
of a plant, such as we set out to give, would be complete at this point. 
However, in view of the extraordinary discrepancy between the somatic 
and semi-reduced chromosome number in the majority of roses on the one 
hand, and of the regular haploid and diploid relationship in those of Rosa 
arvensis and the average phanerogam on the other, the facts must be 
emphasized by figures of a somatic mitosis in that rose. 
PI. IX, Fig. 21 shows a somatic nucleus in late prophase, exhibiting very 
clearly the splitting in preparation for the ensuing division, whilst Fig. 22 
shows a typical equatorial plate. In both, without the faintest possibility 
of equivocation, the somatic number is 14. In other words, the haploid 
number for Rosa arvensis is 7 and its diploid 14, these numbers bearing the 
normal relation to each other. 
(b) Rosa rugosa , Thun. 
Rosa rugosa, an occasional escape or covert shrub in this district, 
displays essentially similar features to those described for R. arvensis , 
its diploid number likewise being 14 and its haploid 7. Only one fact 
seems worthy of special mention in connexion with this rose, and that is 
the very considerable quantity of granular chromatin matter present in the 
nucleus during diakinesis. Otherwise it is very ordinary in its behaviour. 
Under these circumstances, to avoid unnecessary duplication, we do not 
supply a series of figures for it, but content ourselves by inserting a good 
example of a homotype anaphase (PI. IX, Fig. 17) to illustrate its perfect 
regularity at a stage when most rose forms exhibit their worst behaviour. 
(c) Rosa pimpinellifolia, L. 
With this form we include the hispid glandular peduncled R. spinosissinia 
as well as a tall sterile bush indistinguishable otherwise than in its 
stature and sterility from R. pimpinellifolia. All three forms, in contrast 
to the irregular tetraploid Villosae, appear, save in one slight detail in the 
sterile type, as wholly normal tetraploids. 
