THE HORTICULTURAL SOCIETY OF NEW YORK 
and more slender and the leaves were somewhat smaller. The 
difference in the growth of first and second bulbs is well shown 
in Nos. 1 and 2 of Plate 38. 
Occasionally the first bulbs produced two flowers, as is shown 
in No. 1; this was due to the blooming of a small bulb, which at 
the time of planting was situated within the outer scales of the 
first bulb. Such flowers were always of small size. 
The four plants shown in Plate 38 were grown from sister bulbs 
weighing 29.5, 5.45, 2.75, and 1.7 grams. This series was photo- 
graphed to illustrate the principal types of growth observed. Nos. 
1 and 2 show the relative size and vigor of the growth from first 
and second bulbs. No. 3 shows a blind tulip and No. 4 shows 
the development of a scale leaf only. 
The blooming of bulbs of Cottage Maid was less pronounced 
than in Rose Grisdelin. Fewer second bulbs bloomed and no 
bulbs of smaller size bloomed. 
As would perhaps be expected, the greater proportion of bulbs 
that died or that developed a scale leaf only was among bulbs of 
smaller sizes. The total number of blind bulbs was low—6 of 
Rose Grisdelin among third bulbs and 2 of Cottage Maid among 
first bulbs. 
It may be said that in planting for special display one would 
scarcely ever select any bulbs of the size of the second bulbs used 
in the experiment. Probably no bulbs of smaller size than the 
first bulbs would be sold for planting by any commercial dealer. 
The practical methods of increasing stock depends on the vege 
tative multiplication of bulbs giving sister bulbs of various sizes 
and the subsequent growth of the smaller of these bulbs without 
their blooming or splitting up into smaller bulbs until bulbs of 
large size are formed. It would seem that the blooming of bulbs 
of smaller size is undesirable in that it may retard the develop- 
ment of bulbs of second rank to the size of first rank. On ac 
count of the smaller size of the flowers produced and the some 
what greater uncertainty that flowers will be produced, it % 
scarcely practicable to select any but largest or first bulbs for 
planting. It is planned to determine the further performance of 
the bulbs of second and lower ranks which produced flowers 
the experiment. A. B. Stout 
206 
