DR. M. T. MASTERS ON THE PASSIFLORACEZÆ. 621 
2. Relative invariability, a point on which deservedly much stress is laid, is to a great 
extent implied in the Table showing the relative frequency of occurrence of certain 
characters (Table 1)—but not wholly so; for instance, the number of petiolar glands and 
of rings of the corona is subject to variation even in the same species. It is a rare 
thing for them to be entirely absent; but it is common for them to be variable. Con- 
staney of occurrence and invariability of form or arrangement do not necessarily go 
together. Nevertheless, as a general rule, it may be stated that those characters placed 
high in the list on account of their frequent occurrence, are also those which are least 
liable to variation in individual plants, or even species. "Thus the one-celled ovary and 
the parietal placentation are not only characters of general occurrence, but they vary 
comparatively little in form in the individual members of a species. On the other hand, 
the corona and the petiolar glands, which are almost peculiar to Passifloracee, and 
which are therefore placed at the bottom of the list of characters, from the infrequency 
of their occurrence in other orders, are very liable to variation in individual plants. 
3. Developmental characters.—A glance over the characters alluded to in the foregoing 
lists will suffice to show that some of them are * congenital," and others “acquired.” 
Of course there is only a difference of degree between congenital and acquired characters; . 
nevertheless the difference is an important one, and one readily recognized. 
The acquired characters are necessarily subsequent to the others in the time of their 
appearance; they belong, as it were, to another stage of growth. 1f the illustration may 
be permitted me, I would compare the congenital characters to the “carcass” of a 
house, the acquired ones to its fittings. 
The one set of characters are peculiar to, and perhaps different in, each individual, 
according to circumstances; the others (though also susceptible of modification) are 
common to all, or to a large number. Now, in the illustrations already given, it will be 
seen that those characters which are the most important in point of frequency of occur- 
rence are precisely those which are congenital. 
Again, it may be seen that those characters which are acquired, those which make 
their appearance later in the course of development, are those upon which stress is laid 
in distinguishing the lower groups, such as genera, sections, and species, one from 
Their use is in analysis chiefly, as that of the congenital characters is in 
another. 
synthesis. Moreover these acquired characters are taken from organs which, speaking 
generally, are those concerned in the carrying on the physiological actions peculiar or 
special to individual plants. But congenital characters are not merely of use synthe- 
tically, they may occasionally be used analytically, when they afford the means of dis- 
criminating, not only between large groups, but also between the members of the group 
itself. Thus the presence or absence of stipules affords congenital characters available for 
distinguishing both large and small aggregates from each other. At the same time the 
stipules are physiologically important as protectors to the young buds, or, in some cases, 
as secreting organs*. 
* From their frequent very large comparative size in the young bud it is fair to infer that their office is one of 
greater importance in that stage of development than it is subsequently, when the leaves are fully developed. | 
