by Bob Harms ()

Berberis trifoliolata var. glauca I. M. Johnston

A variety Mahonia trifoliolata var. glauca was proposed by I. M. Johnston (J. Arn. Arb. 31, 1950), with the description:
A varietate typica differt foliis opacis glaucis minutissime abundantissimeque papillatis.

[My translation:
    It differs from the typical variety in leaves that are opaque, glaucous, abundantly and minutely papillose.]


Papillose leaflet surfaces from the Davis Mountains in West Texas (Jeff Davis Co.)
adaxial on the left; abaxial on the right

Kenney & Peebles, The Flora of Arizona (1960), note that

The Arizona plants presumably all belong to Mahonia trifoliolata var. glauca Johnst., described as differing from typical B. trifoliolata in the opaque glaucous, minutely papillate leaves.
Ahrendt's 1961 monograph incorporated the new variety into his key, with no mention of glaucous:

(602) Leaflets above slightly lustrous green, and epapillose
   ... 602a. M. trifoliolata var. trifoliolata
     [For surface images of this type.]

(602) Leaflets above conspicuously grey-green, and very minutely and closely papillose
   ... 602b. M. trifoliolata var. glauca


Papillose leaves from the Davis Mountains (Jeff Davis Co.)
(Magnified surface views above.)

Later, this variety seems to have taken on a life of its own on the basis of the its unfortunate name 'glauca' without consulting Johnston's article. Had the variety been named 'papillata' much of the subsequent discussion might not have occurred. For example, Correll & M. C. Johnston 1970:

Those plants with glaucous foliage are usually segregated as var. glauca (I.M. Johnst.) M.C. Johnst. (Mahonia trifoliata var. glauca I.M. Johnst.)
The name thus became:
Berberis trifoliata var. glauca (I.M. Johnst.) M.C. Johnst.

But M. C. Johnston only revised (updated) the name, replacing Mahonia with Berberis, and did not actually propose a new description. Indeed his qualification 'usually' may simply reflect observed practice and not approval of the simplification by omition of 'papillose' texture.

The name revision did not, however, constitute a valid name publication under the International Code of Botanical Nomenclature, Article 33.3, since it lacked "full and direct reference given to its author and place of valid publication, with page or plate reference and date."

In his 1990 The Vascular Plants of Texas: A List, Up-Dating the Manual of the Vascular Plants of Texas (2nd Ed.) Johnston presents a revision (p. 31):

Berberis trifoliolata Moric. My invalid combination with Berberis (in the Manual, but lacking literature citation) of Mahonia trifoliolata var. glauca I. M. Johnst., has not been validated. The plants in the extreme eastern part of the range of agarito (southern and central Texas) have more less green foliage. Clinally, westward, the leaves become more glaucous. But the absence of a sharp distinction and of any other correlated character-state weakens the case for recognition of a "variety."
In 1992 J. E. Laferrière applied M. C. Johnston's name with a somewhat refined definition, closely following the original:
Var. glauca (I. M. Johnst.) M. C. Johnst. ex Lafer. (glaucous) --Leaflets glaucous, minutely papillose (under 30X magnification). [B. trifoliata Hartweg. ex Lindl.]. --Slopes and mesas: Santa Catalina and Rincon Mts. of e Pima Co.; 900-1050 m (3000-3500 ft); Mar-Apr; w TX and NM, s to Hgo., Mex.
Apparently the fact that the date and journal source appear on the pages of this name — Journal of the Arizona-Nevada Academy of Science, Vol. 26 (1), p.4, 1992 — finally give it validity, if not acceptance.

Whittemore's 1997 (FNA) treatment of B. trifoliolata objected to M. C. Johnston's name as illegitimate along with a rejection of the status of 'var. glauca' on empirical grounds:

The illegitimate name Berberis trifoliolata Moricand var. glauca (I. M. Johnston) M. C. Johnston has been used for plants with very strongly glaucous leaves. Weakly and strongly glaucous plants are often found in the same population, however, indicating that they are not distinct varieties.
Whittemore's observation is more or less valid for several of our Central Texas populations (cf. Berberis Leaf Surfaces in Central Texas) — the variation even includes nonglaucous forms. But, as should be clear from the above discussion, the variety was not proposed on the basis of glaucous leaves, but rather on its very strikingly different papillose adaxial leaf surface.

Both I. M. Johston's and Laferrière's varietal names are currently listed by the USDA Integrated Taxonomic Information System as synonyms for Mahonia trifoliolata and as 'Current Standing: not accepted – synonym.'

Mahonia trifoliolata (Moric.) Fedde var. glauca I.M. Johnston =Mahonia trifoliolata
Berberis trifoliolata Moric. var. glauca (I.M. Johnston) M.C. Johnston ex Laferrière =Mahonia trifoliolata
The expert given for rejection of this variety is John Kartesz, Biota of North America Project. But apparently the basis for the decision is not available — perhaps the influence of Whittemore's spurious rejection noted above.