105 



quently yield, among nearly-rekted species, tlie most tangible 

 criteria, or afford a ready means of recognising affinities. More 

 tlian that, tliey must be available if we are to arrive at a comj^lete 

 knowledge of a species. For that purpose, to speak only of the 

 :flowering plants, leaf and stem and root are organs as essential 

 as flo^yers and fruits. The ideal specimen would represent the 

 plant in its normal state, with all its parts, and the parts chosen 

 so as to show their arrangement as well as the principal stages 

 of development and the range of their variation. Though tliis 

 ideal_ may not be easily attained it should be kept in mind as a 

 guiding principle in selecting specimens. 



The collector should make it a rule, to give time to selection 

 and subsequent preservation rather than to aim at a great number 

 of specimens, which will probably be less satisfactory and may, 

 in the end, have to be thrown away. As long as the collector 

 luis to deal with small plants which do not exceed the dimensions 

 of his collecting outfit, and can tlierefore be taken up entire, he 

 will only have to look for healthy average individuals which bear 

 ilow^ers or fruits or spores, and he may add to them some in 

 earlier stages of development. If they are not much larger, and 

 can be doubled up, he may still do the same. But if they are too 

 large to he folded — as many lierbs, ferns, certain seaweeds, 

 most shruhs and all trees — he will have to resort to a selec- 

 tion of sections to fit his paper or his bottles, and he may find 

 it difficult to cut pieces not too large and j^et representative 

 of the j)lant. In those cases it will be best to choose, in the first 

 place, a piece exhibiting the organs of reproduction in theii 



raitural connection with as much of the vegetative parts as 



possible, and then to supplement it, if necessary, by othei 

 .sections which contain the missino* parts, care being taken that 



om tiie same 



om 



Siicli detaclied specimens must be so marked tliat their connection 

 cannot be lost or doubted. But even so, some characters, such as 

 habit, dimensions, etc., will not be shown by the sijecimens, 

 "however well selected, and must therefore be supplemented by 



notes or sketches. 



The following essential points may be summarised as rules for 



.selection. 



1. Select specimens so that they shall be as representative of 



■the plant as possible. 



2. Do not collect specimens icitliout organs of reproduction, 

 •exicept in very special cases, such as in those of prominent 

 aiembers of the vegetation of a new country, of plants of special 

 biological or economic importance, or of plants growing at the 

 outposts and limits of vegetable life. 



3. Select specimens so that they exhibit their organs of repro- 

 duction in connection with as much of the vegetative parts as 

 l^ssible, and, unless unavoidable, do not detach flowers and fruits 

 except with the object of supplementing the specimen. 



4. In the case of small plants, take up whole individuals with 

 fhe root, selecting the most typical of their kind. 



5. In the case of larger plants, cut specimens, in the first 



1 



