Marisora alliacea

Marisora alliacea (Cope)

Middle American Four-lined Skink

Fig. 10A, B, C

Mabuia alliacea Cope 1875:115 (no holotype designated, but Dunn 1936:539 and Cochran 1961:125 b listed USNM 30619–20 as syntypes; no type locality given, but Taylor 1956:298 gave “ Costa Rica ” “low country” [= Caribbean lowlands] as the type locality).

Marisora alliacea: Hedges & Conn 2012:119; Sunyer et al. 2015:384; HerpetoNica 2015:220; Sunyer et al. 2016:1052.

Mabuya unimarginata complex: Pinto-Sánchez et al. 2015:195 (in part).

Diagnosis. Marisora alliacea is a long-limbed, relatively large species of the genus characterized (data from five males and nine females [marked by an * in Appendix 1], plus data from Taylor 1956, where noted) by (1) maximum known SVL in males 79.0 mm; (2) maximum known SVL in females 90.3 mm; (3) SW 2.6–4.7% SVL in males, 2.4–5.1% in females; (4) HL 17.7–22.8% SVL in males, 16.0–22.8% in females; (5) HW 11.3–19.2% SVL in males, 11.3–17.3% in females; (6) EAL 1.1–2.0% SVL in males, 1.0–2.4% in females; (7) Toe IV length 11.4–13.3% SVL in males, 9.5–12.6% in females; (8) prefrontals one per side; (9) supraoculars four per side; (10) supraciliaries 4 per side; (11) frontoparietals one per side; (12) usually sixth supralabial below orbit (73.8%), fifth below orbit in 26.2% (includes our data and that from Taylor); (13) nuchal rows one per side; (14) dorsals 51–60 (53.6 ± 1.4) in males, 50–60 (56.0 ± 2.5) in females (includes our data and that from Taylor); (15) ventrals 59–62 (60.6 ± 1.3) in males, 56–65 (58.5 ± 3.3) in females; (16) dorsals + ventrals 112–115 (113.6 ± 1.1) in males, 107–123 (114.0 ± 4.6) in females; (17) scales around midbody 28 in 48.5%, 26 in 40.0%, rarely 27 or 29 (includes data from Taylor); (18) Finger IV lamellae 12–15 (13.4 ± 1.1) per side in males, 13–15 (13.4 ± 0.7) in females; (19) Toe IV lamellae 15–18 (16.6 ± 1.1) per side in males, 15–18 (15.6 ± 1.3) in females; (20) Finger IV + Toe IV lamellae 28–31 (30.0 ± 1.2) per side in males, 28–31 (29.0 ± 1.4) in females; (21) supranasals only occasionally (17.0%) in medial contact, thus frontonasal-rostral contact in 83.0% (includes our data and that from Taylor); (22) prefrontals not in contact medially; (23) supraocular 1-frontal contact absent; (24) parietals in contact posterior to interparietal; (25) pale middorsal stripe absent; (26) distinct dark brown to black dorsolateral stripe or lines present above a pale brown to cream dorsolateral stripe; supplemental thin dorsal stripes or lines present, those supplemental lines sometimes broken into dashes; (27) dark brown lateral stripe present, broad (3–4 scale rows high), at least in shoulder region; (28) distinct white lateral stripe present; (29) palms and soles dark brown to black; (30) total lamellae for five fingers 46–51 (47.8 ± 2.2) in males, 42–53 (47.6 ± 4.2) in females; (31) total lamellae for five toes 53–62 (56.3 ± 4.4) in males, 52–55 (53.6 ± 1.1) in females. In addition, this is a long-limbed species with the combined FLL + HLL/SVL 62.5–74.6% in males, 58.0–67.6% in females (includes data from Taylor), and 2 chinshields contacting infralabials in 84.4% and one in 15.6% (Table 3).

Marisora alliacea is a member of the M. alliacea Group of Middle American mabuyids and is apparently most closely related to M. roatanae (Fig. 3) [but tissues not available for the also Caribbean lowland M. magnacornae]. Marisora alliacea differs from M. roatanae in having longer limbs (FLL + HLL/SVL 62.5–74.6% in males and 58.0–67.6% in females versus 53.5–58.4% and 47.8–57.7%, respectively, in M. roatanae), having 26–28 scales around midbody in 88.5 % (versus 30–32 in 76.7% in M. roatanae), and having the frontonasal contacting the rostral in 83.0% (versus that contact absent in M. roatanae). Marisora alliacea differs from M. magnacornae by the combination of having the sixth supralabial below the orbit in 73.8% (versus fifth supralabial below orbit in 85.5% of M. magnacornae), having the frontonasal contacting the rostral in 83.0% (versus frontonasal separated from rostral in 96.9% of M. magnacornae), and having 28 in 48.5%, 26 in 40.0%, or rarely 27 or 29 scales around midbody (versus 30 scales around midbody in 93.1% and 28 in 6.9% of M. magnacornae). Marisora alliacea differs from all remaining Middle American Marisora species of the M. alliacea group studied herein by having long limbs (FLL + HLL/SVL 62.5–74.6% in males and 58.0–67.6% in females versus <60% in males and <58% in females of those species). Marisora alliacea differs from the extralimital M. pergravis in having dark dorsolateral stripes (versus those stripes absent in M. pergravis). Marisora alliacea has been confused with M. unimarginata of the M. unimarginata Group of the genus in several poorly documented, incomplete studies. Marisora alliacea differs from M. unimarginata in having the frontonasal contacting the rostral in 83.0% (versus frontonasal separated from rostral by supranasal medial contact in all M. unimarginata examined), having two chinshields per side contacting infralabials in 84.7% (versus one chinshield per side contacting infralabials in 82.9% in M. unimarginata), and in having dark brown dorsal stripes or lines, or dashes suggesting lines (versus dark brown to black dorsal spots present in M. unimarginata). Marisora alliacea is known to differ from the extralimital and poorly known M. berengerae (incomplete morphological data available only from the unsexed holotype) of the M. unimarginata group only from genetic data; furthermore a large geographical hiatus occurs between those two species.

Distribution. Marisora alliacea is a Caribbean versant lowland species that is known to occur from southeastern Nicaragua to Bocas del Toro Province in northwestern Panama (Fig. 6). All specimens of this species we examined are from below 300 m elevation. The specimens examined by Taylor (1956) also are from localities below 300 m elevation. Savage (2002) plotted numerous Costa Rican Caribbean versant localities that are certainly from higher elevations but did not provide supporting locality data or museum specimen numbers for those localities.

Remarks. Our genetic results (Fig. 3) support the morphological studies of Taylor (1956), Hedges & Conn (2012), and this study that Marisora alliacea is a valid species. Pinto-Sánchez et al. (2015) also recovered M. alliacea as a separate clade in their phylogenetic analysis, but, oddly, did not recognize that species or comment on that result.

A tissue sample from northeastern Costa Rica presumed to be of this species was sequenced by Miralles et al. (2009b) and included in studies by Miralles & Carranza (2010), Hedges & Conn (2012), Pinto-Sánchez et al. (2015), and in the current study. Unfortunately, the voucher specimen for that sequence apparently was not collected (Miralles & Carranza 2010:861). A second Marisora alliacea sequenced for this study from southeastern Nicaragua does have a voucher (MVZ 269259) that upon examination proved to be typical in morphological characters with M. alliacea.

The reproductive data presented by Goldberg (2009; as Mabuya unimarginata) are not informative at the species level because the summary data provided therein are a complex of Marisora alliacea, M. brachypoda, and M. unimarginata.

Images of Marisora alliacea are in HerpetoNica (2015), Köhler (2003, 2008; both as M. unimarginata from Bartola), Sunyer et al. (2016), Taylor (1956), and Vences et al. (1998; as M. unimarginata). Fitch (1985) reported three M. alliacea from Costa Rica gave birth to 2, 2, 1 young (no dates given), but Fitch (1975) reported that one of those females was found in March.