FLORA OF WASHINGTON AND VICINITY. 231 



wards, if additional specimens are obtained from other localities, they 

 can occupy the upper half, with the labels under them in the middle of 

 the sheet. 



In fastenin<; down the labels it is not best to gum the entire surface, as 

 they will then roll up, warp, and assume a wry position which can never 

 be cured afterwards. This can, it is true, be prevented by immediately 

 putting that corner of the sheet into a clamp and leaving it there till dry, 

 or by using heavy weights, but this is generally difficult or impracticable 

 where a large number of labels are to be put down at one time. By 

 gumming only a narrow portion of the upper margin of the label there 

 will be no warping, and I recommend this plan. It is proper, however, to 

 state the objection to it, which condems it in the eyes of some. This is, 

 that in handling the plants one is apt to take hold of the loose portion of 

 the label and tear it otf. I have never yet torn one, and do not think the 

 objection serious, but at least it need not be, if the sheets are manipu- 

 lated with the thumbs and near the middle, in the manner described a 

 few pages back. If pains are taken in putting down the label to have 

 its outer edges fall a trifle inside those of the sheet, there will be no 

 danger of ever taking hold of the label. 



5. — Caee of duplicates. 



Some botanists pay little attention to their duplicates, arrange them 

 in no definite order, keep them in parcels, each summer's collection by 

 itself, or in other unsystematized wa,A s, and depend upon memory to 

 hunt out anything they may want to iind. This is in a high degree rep- 

 rehensible, and really occasions great loss of time. Others arrange 

 them in the alphabetical order of the genera, which is much better, but 

 is not to be recommended. It is best to arrange them carefully, accord- 

 ing to the natural system, the same as the herbarium. 



How to label the cases of so shifting a mass has.been a serious difficulty. 

 I have heard very few plans of doing this suggested, and I think nearly 

 all botanists leave them without labels and depend upon memory to dip 

 in wherever they think their i^lant is. I will give my own method, which 

 has worked admirably, and which eminent botanists have admired and 

 expressed an intention to adopt. 



Strips of white paper, 19 inches in length, are cut of two widths, one 

 kind 2 inches wide, the other 1 inch. The former are used for genus 

 strips, the latter for species-strips. Every genus is furnished with one of 

 the wider sort and its name is written across one end, which projects far 



