366 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY, 
prevailing in his trials with the ' French ' and ■ Doucin ' stocks are 
hardly applicable to this country, though the account is interesting. 
It seemed quite clear, then, that the present work would have to 
start from the very beginning in its method of classification, and that 
it was highly desirable once and for all to put on record detailed 
descriptions of the various types before attempting to judge their 
merits. 
IV. Beginning of the Work. The Collection. 
Our investigations began in November 191 2. They were under- 
taken in collaboration with Prof. B. T. P. Barker, M.A., of the Agri- 
cultural and Horticultural Research Station of Bristol University at 
Long Ashton, Somerset. Long Ashton was to specialize, in the first 
place, upon the ' Free ' and ' Crab ' Stocks, whilst East Mailing dealt 
with the ' Paradise ' forms. The work at East Mailing was initiated 
by Mr. R. Wellington, the Director of the Station. The original 
scheme and the subsequent isolation of Paradise types was primarily 
his work. On his shoulders fell the brunt of the initial difficulties. 
When war broke out in August 1914, Mr. Wellington joined the 
forces and asked the present writer to continue the investigations. 
Though the work had not then been under weigh long enough to make 
possible the detailed comparative study of types, and though further 
types were yet to appear, Mr. Wellington left behind invaluable notes 
and records for the use of which he gave free permission. His only 
anxiety was for the continuation of the investigations and the issue 
of the report thereon. At the same time the continuity of the whole 
experiment was safeguarded by the retention of Mr. J. Amos, the 
foreman-assistant and recorder, whose able and untiring services 
bridged over a critical period and who rendered the greatest help in 
collecting the facts for the present report. In addition, the writer 
is indebted to Mr. H. Wormald of Wye College Research Department, 
who has frequently given his aid and advice on botanical matters. 
The majority of nurserymen in the British Isles were approached 
and asked to send samples of the various kinds of Paradise used and 
grown by them. The invitation, in most cases, met with a ready 
response, and a considerable collection was thus obtained. As it be- 
came evident that many nurserymen " bought in " their stocks from 
abroad instead of raising them personally, the collections were added to 
by further samples from representative nurserymen in Holland, France, 
and Germany. In all 71 collections * were obtained from 35 sources. 
The British sources numbered 29, f the French 3, the Dutch 1, and the 
German 1. This last source, however, furnished 20 collections. Each 
collection was numbered as it arrived, and was planted out separately. 
Most samples contained twelve stocks purporting to represent some 
specified variety, named by the nurseryman. Three stocks out of 
* Including one true collection of ' Northern Spy ' (not a Paradise type), 
f The thirty British sources were from fifteen different counties, either 
noted for fruit-growing or nursery work. 
