188 HOW TO STUDY PLANTS. [LESSON 31. 



however, to split down the corolla tube and note that the insertions 

 of these four stamens are alternate with the blended petals. 



Under 3, of course, we select " * * With ordinary green herb- 

 age " ; and under this the second choice, "!--- Corolla irregular : 

 stamens four." 



541. Four choices are now presented to us, dependent upon the 

 character of the ovary. A surface examination of the ovary, 

 accompanied by a careful cross-section, will show an entire ovary, 

 which is two-celled, and contains very many small ovules. Such a 

 combination of characters can only be found under the first divis- 

 ion, and accordingly we select the order SCUOPHULAKIACEJE, the 

 fifty-sixth of the Manual, p. 271. 



542. We find here 16 genera, grouped into two sections, I. and 

 II. An examination of a fully developed bud will at once show 

 that the " upper lip of the corolla is external in the bud." and that 

 our plant belongs to one of the first 8 genera. 



The form of the corolla (not saccate or produced at base), no 

 rudiment of a fifth stamen, the solitary pedicels in the axils of the 

 leaves or bracts, and the splitting capsules, all point to the third 

 group, marked with three asterisks. 



Under this there are three sets of characters, marked by daggers, 

 under the first of which our plant must belong, for it has a barely 

 5-toothed calyx, a decidedly bilabiate corolla, and four stamens. 

 Hence it is to be referred to the genus MIMULUS, the sixth of the 

 order, and a perusal of the generic characters confirms our choice. 



543. Turning to page 279, we find the seven species of Mimulus 

 arranged in two groups, and as our plant is ' * * Neither viscid 

 nor glandular," we select the second. 



The yellow corolla, and oblique calyx with unequal teeth, at 

 once refer us to the group marked > -i , under which we find a 

 choice to be made between two species. 



It is very probable that a reading of the specific characters will 

 lead to a selection of M. luteus, for it is exceedingly common, but 

 allowance must be made for its extreme variability. 



544. All the characters displayed by a variable species cannot 

 be compressed into a single description, and hence, if the Mimulus 

 in hand satisfies M. luteus more nearly than any other species, it is 

 safe to call it by that name. 



545. After several analyses of this kind, the student will be able 

 to pass rapidly over most of the preliminary steps, and should be 



