On the identity of Helianthemum mathezii and H . pomeridianum ( Cistaceae )

The original description of Helianthemum mathezii regarded the species to be a therophyte. However, the detailed observation of the holotype of H. mathezii, as well as newly collected specimens from the type locality, does not support its condition of annual plant. Further study has led to the conclusion that all these plants can readily be identified as H. pomeridianum; the descriptions of H. mathezii and H. pomeridianum are equivalent except for the habit, being the former annual and the latter suffruticose. We show in this study that H. mathezii is to be considered a synonym of H. pomeridianum, and report a new locality of H. pomeridianum in Morocco. Also, we provide the first chromosome count for this species (2n = 22) which emphasizes the close relationship of H. pomeridianum to H. subgen. Plectolobum.


INTRODUCTION
described H. mathezii Dobignard -Cistaceae Juss.-from a single locality from the Moroccan High Atlas -Taroudant province-as a 20-35 cm tall, much-branched, and glandular-hairy annual plant.This author stated that, despite its apparent morphological resemblance to H. salicifolium (L.) Mill.(H.subg.Helianthemum), H. mathezii should be ascribed to H. sect.Atlanthemum (Raynaud) G. López & al. (H. subg. Plectolobum Willk.)due to its glandular and prominently 5-7-veined inner sepals that resembled those of H. sanguineum (Lag.)Lag.ex Dunal. Dobignard (2009) also reported that this species inhabited wet travertine deposits accompanied by Feeria angustifolia Buser and Euphorbia rimarum Coss.& Balansa, among other species.Thus described, H. mathezii constituted a very interesting species inasmuch as it was the only Helianthemum species inhabiting a wet habitat (cf.Arrington & Kubitzki 2003), and the only annual species in the genus with such a restricted distribution (cf.Proctor & Heywood 1968;Greuter & al. 1984), further included in the checklist of North African endemic plant species (El Oualidi & al. 2012).Consequently, as part of a comprehensive phylogenetic study of the genus Helianthemum (Aparicio & al. 2017), we considered of interest to perform a new collection and a detailed study of H. mathezii.

MATERIAL AND METHODS
To do so, we visited the type locality of H. mathezii, at "piste d'accès au plateau du J. Tichka alt.c. 1100 m, 30°47'18 N, -8°40'73 W" (Dobignard 2009), to survey the species.For the chromosome number determination we collected matured seeds that were disposed for germination in Petri dishes.Then, root tips were immersed in chilling water at 0 ºC for 24 h, then fixed in 1:3 glacial acetic acid and absolute ethanol for 3 h.They were finally stained in alcoholic hydrochloric acid-carmine for 48 h and squashed in 45% acetic acid.
No type specimen of H. pomeridianum Dunal has been observed.Nevertheless, H. pomeridianum has been many times collected by different authors in its Algerian locus classicus -circa Oran, Djebel Santo-, and vouchers are available on-line at the Muséum Nationale D'Historie Naturelle -MNHN-.We list the studied ones in appendix 1, all identified as H. pomeridianum.

RESULTS AND DISCUSSION
During our field trip to the type locality, we found a population integrated by about 30 individual plants readily identifiable as H. mathezii, other than for the fact that these specimens, albeit having slender branches, were perennial suffruticose plants (fig.1).Moreover, they were located in a sunny dry area on the top of a calcareous travertine deposit accompanied by Micromeria hochreutineri Maire and Lavandula multifida L. rather than in the wet vertical run-off area of the same travertine outcrop, where Feeria angustifolia and species of Euphorbia L., Campanula L., and Erodium L'Hér.were common.
The detailed observation of the holotype of H. mathezii (fig.2), as well as the figure no.7 of the original description of the species (Dobignard 2009), does not provide conclusive evidence that H. mathezii is a therophyte since the root system cannot be observed, a fact clearly evident in the specimens of H. sanguineum and H. salicifolium therein included for illustration -not shown here-.Moreover, the analysis of the Dobignard's material and our newly collected plants, following Quézel &Santa (1962) andFennane &al. (1999), has resulted in that all these plants can be identified as H. pomeridianum.Further study has also shown that the description of H. mathezii is equivalent to that of H. pomeridianum except for its habit -the former annual, the latter suffruticose-since both descriptions depict glandular plants which have: 1) slender branches bearing very shortly petiolate, exstipulate, flat, ovate-lanceolate leaves markedly veined beneath -falsely stipulate because of the presence of small axillary crowded leaves-; 2) very loosely arranged flowers in glandular terminal cymes composed by a few pedicellate and bracteate solitary flowers; 3) inner sepals 4-6 × 2 mm, striate with prominent veins; 4) stamens numerous; 5) pubescent ovary bearing a short capitate style; 6) capsule shorter than the inner sepals, glabrous, with ciliate margins at the apex (Dunal 1847;Willkomm 1856;Dobignard 2009).Overall, these arguments show that H. mathezii is a synonym of H. pomeridianum, and that both Dobignard's specimen -G 00441364-and our own collection -SEV 286762 (Morocco: Taroudant, between Sidi Abdellah Oussaid and Alegjane, 1100 m a.s.l., 30°47'10.Helianthemum pomeridianum is a rather poorly known North African endemic species (El Oualidi & al. 2012), which was originally described, and many times collected by different authors, from around Oran -Djebel Santo, Santa Cruz-in Algeria.To date, H. pomeridianum has also been recorded in Melilla (Caballero 1917), the Trara Mountains, also in the northern Algeria (Medjahdi & al. 2009), and in a few far away locations in the Moroccan High Atlas area, between Taroudant and Marrakech -RAB 078906, MA 472377 (P 04729082 photo!), MGC 80897 (MA 321784-1!)-(Förther & Podlech 2002;GBIF 2016   Finally, we were unable to extract DNA from a leaf sample of the type specimen of H. mathezii provided by herbarium G despite we followed two standard -and slightly modified-DNA extraction protocols: Invisorb Spin Plant Mini Kit -Stratec-and Isolate II Plant DNAkit -Bioline-.Consequently, this accession could not be included in the phylogenetic analysis of Helianthemum, but in the phylogenetic analyses our samples collected in type locality -SEV 286762-clustered with other Moroccan accessions of H. pomeridianum Dunal and maintained the supported sister-taxa relationship between H. pomeridianum and H. lunulatum (Aparicio & al. 2017).