328 XI-:"' EXGLAXD TREES IX WIXTER. 



tlio Ash are not oovltlmI l.iy persistent bases of leaf-stalks and there 

 are 2 or more jiairs of scales to the leaf biuls. we therefore pass 

 to 14. The relatively small size of the Inicls ami their freedom 

 from 5tiel<iness takes us to 15 where the number of the bundle- 

 scars and the other characters given show us that we have in hand 

 a twig ot one eif the Ashes. AVe now turn tii Xo. ItU and confirm 

 our determination ot the genus liy reading the general description 

 of the Aslies. To lind mtt which Ash we are dealing with we 

 procede with the key of the Ashes and. knowing that tlie leaf-scars 

 in. our specimens are deeply concave on their upper margins, we 

 conelnde that we have tJie A\liite Ash (Fm.vinus americana). 

 At page obij we find a detailed description with photographic 

 illustrations of this species and may learn the winter characters 

 of the tree not already given in the key. If tlie description and 

 ]diotographs do irot correspond to the tree under investigation, we 

 know that we have gone astray at si.uue poiirt in the key. and 

 turning Ijack \\'e I'cpeat the analysis taking if need lie the other 

 alternative of a jiair where the choice had Ijcen doulitftd. 



Sontetimes a tree is varialde in the characters used in the key. 

 Tlius the (Iiestnut lias terminal bttds sometimes present on the twigs 

 though they are generally absent. In sttcli cases, however, and 

 wliere there is a legitinuite doubt as to wdietlier the tree should be 

 placed in the first oi' the second group, it has generally Ijeen placed 

 in lidth so that either of tlie two (hnirts should lead tn the correct 

 name. The determination of the presence or absence of the tei'- 

 minal buil is jierhaps the greatest jiitfall likelv to be found in the 

 use ot the key. littt should give little ditficulty if the discussion in 

 tile introductiiiii is uiiderstooil and the terminal scar is looked for 

 with the aid of a haml-lens. 



