102 Mr P. Sewell on the Morpliology and [sess. li. 



resemblance to Veronica, that the upper petal in Plantcujo 

 must be regarded as the result of the union of two; but the 

 nervation in this case does not come to our aid, as the veins ^' 

 are almost entirely wanting in the membranous petals. 



If, then, we regard Plantaginacete as thus related to 

 Scrophulariaceae, they should be grouped together, as indeed 

 is the case in the rough classification by Goebel. 



On the contrary. Sir Joseph Hooker, in his appendix to 

 " Le Maout and Decaisne," has placed the order Plantaginacese 

 apart as an anomalous order of Series II. 



This might perhaps be allowed in an artificial classification, 

 Ijut it is not warranted, in view of those affinities, in a pro- 

 fessedly natural arrangement, and this is only one example 

 out of many which might be given showing that the accepted 

 j)rinciples of evolution must enter more largely as a factor into 

 the work of systematic arrangement. 



The tendency to union of the parts of the corolla is seen 

 throughout the order. For instance. Digitalis has the two 

 petals of the upper lobe barely distinguished, whilst at the 

 same time its fifth and upper sepal is always smaller than the 

 other four, suggesting its probable elimination in the future. 

 Gratiola similarly has but a very slight notch in its upper 

 lip to show the two petals of which it is composed. 



The genus Calceolaria, however, offers an exception to the 

 general rule which is laid down by Payer in regard to 

 diagonally arranged sepals occurring in pseudotetramerous 

 flowers, in that it has its sepals placed orthogonally. For it 

 is not likely that the genus, agreeing in other particulars so 

 closely with the mass of the order, is much less nearly related 

 than is Veronica; how then has modification been brought 

 about from an assumed pentamerous form ? 



In the corolla of some of the cultivated species, I found 

 very clearly five bundles passing out from the axis — two in 

 the upper lip, three in the lower, much as in UtricvXaria. 



At times a sport occurs in the lower lobe of the corolla, 

 which seems to indicate a fifth petal, inasmuch as a spur 

 projects from tlic median nerve in the normal pouching of the 

 lower lip. 



In several Stnilh Aniericuu si)ecies, however, I do not find 

 more than four bundles, and in some cases so considerable is 

 the union among the branching's of these bundles that it is 



