372 MESSRS. HENRY AND JAMES GROVES ON THE 
The obvious argument in favour of discarding these more or 
less doubtful names, is that in scientific matters the nearest 
approach to accuracy should be aimed at, and that it is therefore 
better to use a name which, from being accompanied by a full 
and accurate description, or a satisfactory illustration, is known 
to belong to the species, rather than one about which there is a 
doubt. 
On the other side it is urged that a large number of the 
earlier post-Linnean descriptions, and many of the more modern 
ones too, are open to the same objection as are those of Linnzus, 
being of themselves insufficient for identification. "What we now 
consider essential characters were often altogether omitted, and 
variable characters of no import made much of. It would be 
equally necessary to discard these, while to do so would involve 
the changing of an immense number of names. Moreover, in 
the case of most of the more or less doubtful Linnean names, if 
all the evidence is examined closely, a very strong presumption 
can be arrived at as to the plants that were intended. 
We are of opinion that it is the better course to retain the 
names when, although the descriptions are imperfect and of 
themselves inadequate, there are reasonable grounds for inferring 
that they belong to certain plants. 
The following are a few specimen cases in which Linnzus’s 
names, coming under Group 3, have been set aside or, to our 
thinking, wrongly used :-— 
1. Hypericum quapraneutum. Linneus describes this as 
Hypericum floribus trigynis, caule quadrato herbaceo, and cites 
Hypericum Ascyron dictum, caule quadrangulo of J. Bauhin. 
For many years the name was used in this country to denote our 
common four-angled St. John’s Wort, a good distinct species, 
with very conspicuous angles to the stem, occurring almost all 
over Europe. Koch, Fries, Sir Joseph Hooker, and others, 
however, have used the Linnean name for the much less distinct 
and less widely-distributed species, which was known as H. dubiwm, 
Leers, a plant which it is true also has a four-angled stem, but 
the angles are much less strongly marked and are distinctly 
unequal, and its general aspect is much more that of our common 
perforate St. John’s Wort, H. perforatum, Linn. Syme, on the 
other hand, altogether rejects the name on the ground of 
ambiguity. We quote his words as a good example of this 
view :— 
