Panarctic Flora

800602-04 The Hippuris vulgaris aggregate H. lanceolata, H. tetraphylla, H. vulgaris

Geography: Circumboreal-polar.

Notes: Murray, Elven, and Solstad: The three species above share many features and are probably closely related. They differ in leaf shape and number, in stamens, and in site requirements and geographical range. Hippuris vulgaris has about 8 linear or very narrowly lanceolate and acute leaves per whorl, and the single stamen is well exserted from the perianth with filament much longer than the anther. The plant usually grows in freshwater but when found in brackish water it retains the same characteristics. Hippuris lanceolata has about 6 broadly linear or narrowly oblong subacute to subobtuse leaves per whorl, and the stamen is subsessile on the perianth with a filament about as long as or shorter than the anther. The plant occurs both in brackish water (mainly in boreal areas) and freshwater (frequently in arctic areas). Hippuris tetraphylla has about 4 oblong to broadly oblong, obtuse to sometimes emarginate leaves per whorl, and the stamen is as in H. lanceolata. The plant is restricted to brackish waters, mostly to brooks and pools in salt and brackish marshes.

Several authors have included H. lanceolata in H. vulgaris (e.g., McCully and Dale 1961; Hultén 1971a; Löve and Löve 1975a; Scoggan 1979), sometimes as f. litoralis Lindb., and have explained the variation in leaf shape and number as modifications in brackish water. Some authors have even included H. tetraphylla in H. vulgaris (e.g., suggested by Cook 1968). Another common hypothesis has been that H. lanceolata is the hybrid H. tetraphylla x vulgaris, i.e., H. x lanceolata (e.g., Elven 1994; Hämet-Ahti et al. 1998; Mossberg and Stenberg 2003).

We have recently (2009-2010) revised a large material (ALA, AMNH, BG, CAN, DAO, ICEL, O, TRH, TROM) from northern Europe (mainland, Iceland, Bear Island), Greenland, Canada, and Alaska and have reached the following conclusions:

Hippuris lanceolata, H. tetraphylla, and H. vulgaris occur as three distinct, morphologically non-overlapping taxa. We have not found evidence of intermediates or transition among well developed plants (however, juvenile shoots and badly preserved plants may be difficult to identify). Plants in the right reproductive stages invariably have well developed anthers and pollen and well developed (turgid) fruits. We have found no sign whatsoever of abortion. As to the assumed hybrid nature of H. lanceolata, we found that: (1) H. lanceolata is intermediate between the assumed parents in leaf number and shape but not in other features; (2) fruit fertility in H. lanceolata is at level with that of the assumed parents; and (3) the frequent co-occurence of H. lanceolata with its assumed parents is a fiction as a major part of what has been assigned to H. tetraphylla belongs to H. lanceolata, which then is the predominant or sole species in many northern regions. As to the suggestion that H. tetraphylla is a halophytic modification of H. vulgaris, this is countered by the common occurrence of H. vulgaris in waters just as saline as those of H. tetraphylla and sometimes in close proximity with it. We therefore accept three independent species.

The ranges of the three species have largely been misunderstood. All major maps from arctic areas are useless. The maps by, e.g., Hultén (1968a, 1971a) and Hultén and Fries (1986) of H. tetraphylla and H. vulgaris include large amounts of material we assign to H. lanceolata. The same is probably the case with the maps of Tzvelev (1980e, as three species) where we assume that H. vulgaris and perhaps H. tetraphylla has been over-recorded, whereas H. lanceolata probably is heavily under-recorded.

In North America, we find that the majority of the more northern records assigned to H. tetraphylla (and H. vulgaris) rather belong to H. lanceolata. One locality remains for H. tetraphylla in northern Alaska; all the others are transferred to H. lanceolata. Hippuris lanceolata seems to be the exclusive species in the Canadian Arctic Archipelago and is also the only one proved from Greenland (re-determined from H. vulgaris) and Bear Island (re-determined from H. vulgaris). The known range of H. lanceolata is nearly twice as large as that of H. tetraphylla and reaches well outside that of H. vulgaris in some regions, especially in the Canadian Arctic and in Greenland, probably in the Russian Arctic, too.