R N. Salaman and J. W. Lesley 27 



was extremely dry and hot, but twenty-seven of the thirty-one lived to 

 produce tubers so that the mortality after transplanting out was low. 

 In 1912 the whole family available was grown on from tubers. Twenty- 

 six plants were grown on by vegetative propagation and exactly repro- 

 duced the " prostrate " habit of the preceding year. Twenty-one of them 

 flowered, some having heliotrope, some white, flowers. Efforts were taken 

 to secure selfed seed, and this time successfully ; a few set seed naturally. 

 Berries were obtained on six different individuals. A single berry was 

 obtained from a protected and the remainder — about a dozen — from un- 

 protected, generally hand pollinated, flowers. In 1913 the plants were 

 again grown on from tubers and exhibited the same " prostrate " habit 

 as in 1911 and 1912, and a little more selfed seed was obtained. Some 

 of these original plants were again grown on vegetatively in 1914 for 

 the fourth time. Many of them by this time were unhealthy and 

 made poor growth, the leaves failing to develop perfectly. Whilst the 

 healthy plants were again " prostrate " the unhealthy or " miffy " plants 

 as they are termed did not completely reproduce the " prostrate " habit, 

 although it is interesting to note that they retained the characteristic 

 stem structure. This serves to show that it is not safe to classify " miffy " 

 plants without examination of their stem anatomy. In fact, notwith- 

 standing the ease with which the pliable stem of the " prostrate " plant 

 is recognized, experience has shown us that it is best to confirm the 

 observation by an examination of the stems. In most cases sections 

 \vere examined under the microscope, but some only under a pocket 

 lens after staining with phloroglucin. 



The seed arising from the self fertilization of the " prostrates " was 

 sown in 1913, 1914 and 1915, partly at Barley and partly at the John 

 Innes Horticultural Institution, Merton. Some of the resulting families 

 arose from a single ball, others from several balls, but all the families 

 shown separately are independent and not mere fractional or duplicate 

 sowings. The germination was fair, the mortality at planting out not 

 unusually high and the plants quite healthy. 



As Table I shows, nineteen independent families have been raised 

 in three seasons from six individuals of the original " prostrate " family 

 and grown in two different localities. With rare exceptions all the 

 families consisted wholly of "prostrate" plants like the parents from 

 which they arose and presented a striking contrast to adjacent " upright " 

 plants (Plate II, fig. 1). Adding the figures together we have in all : 



"Prostrate" 468. 



"Upright" 3.. 



