Additional data to the herpetofauna of Afghanistan

This study provides observation on 21 species of amphibians and reptiles obtained during various field trips in Afghanistan from 2007 to 2018. The core of this study is distributional data collected mostly by the second author of this paper that were systematically evaluated. Our records show a notable expansion of the current distribution range for several species and highlight the need for more extensive further sampling within the entire country. The first record of Cyrtopodion agamuroides (Nikolsky, 1900) complex for Afghanistan is presented here, increasing the number of herpetofauna in the country to 117 species.


Introduction
Afghanistan is one of the most biodiverse countries at the junction of Central and South Asia, with an extremely varied mountainous and desert topography resulting in numerous habitat types. This fact is also reflected in the composition of amphibians and reptiles, with 116 species (118 with subspecies) in 21 families known to date (Wagner et al. 2016). The herpetofauna of this country is comprised of Palearctic and Oriental elements, two main zoogeographic regions of Eurasia (Sindaco and Jeremčenko 2008;Wagner et al. 2016), which makes Afghanistan a crucial territory regarding research of historical biogeography. Simultaneously, it is one of the least known countries in terms of current biodiversity research due to forty years of continuous war and instability. There are decent foundational knowledge and publications on species diversity and distribution accumulated and published during a forty-year period called "Afghanistan's Golden Age" from about the 1930s to the mid-1970s (see Wagner et al. 2016 for a review). However, data obtained and published after this period are rare and mostly referred to old collections (e.g. Clark 1992;Kuch 2004;Wagner et al. 2016;Jablonski and Lesko 2018;Jablonski et al. 2019). Therefore, the country remains largely unexplored and poorly surveyed. Whereas original distribution data on Afghan herpetofauna obtained in the 21 st century are virtually absent in published literature, to bridge the gaps, we here present additional distribution data recently collected in Afghanistan.

Material and methods
This study compiles records from several independent field observations (own or acquired from other sources) procured from 2007 to 2018 in different parts of the following provinces of Afghanistan: Badakhshan, Baghlan, Balkh, Helmand, Kabul, Kandahar, Paktia, Paktika, Uruzgan and Zabul. Distribution data were taken from 22 georeferenced points (Table 1). Most of the records and photographs came from irregular field trips conducted by the second author (JR) during his non-zoological photographic work in Afghanistan in the period 2008-2011. This author presented some of these records (mainly as photographs or comments in the text) in a small guide (Regan 2017; Fig. 1) with a very general format. Although data in this guide are very interesting and have value, they lack deeper zoological evaluation and contain a number of wrongly identified species. We revised all obtained and available data, corrected species taxonomy and summarised available information on locality, geographic position, date of observation, type of observed habitat and any other pertinent information. Furthermore, Regan´s guide contains additional species that were observed in Saudi Arabia, not Afghanistan and which could confuse further field data compilations for Afghanistan: Trapelus ruderatus (p. 21), Phrynocephalus maculatus (p. 22), and Bunopus tuberculatus (p. 28). Therefore, we excluded these species from the checklist presented below. Other dubious cases or  misidentifications are explained in the main text. All available data presented here were compared with the dataset of localities presented by Wagner et al. (2016) and we made updated maps for certain species using QGIS software (2019). All species or specimens photographs presented here were accessioned into the Herpetology collection of the Florida Museum of Natural History at the University of Florida (FLMNH UF 189172-232; Suppl. material 1).

Results and discussion
We present a checklist of amphibian and reptile species observed and/or photographed in Afghanistan that, in most cases, represent new locality or provincial records for the country. Two families of amphibians and six families of reptiles were recorded with 21 species presented overall in the checklist. An additional species, Cyrtopodion agamuroides (Nikolsky, 1900) complex, is here the first to be presented as another reptile for the species list of herpetofauna in Afghanistan.

Bufotes viridis (Laurenti, 1768) complex
Distribution in Afghanistan. According to Wagner et al. (2016) sented by Buchroithner (1981) that reported these toads from Tila Bay Valley (~37.28N, 73.33E, 4160 m a.s.l.), Wakhan, Badakhshan. This author mentioned an occurrence of B. latastii in Badakhshan, but this species is endemic to western Himalaya and is not known from the Hindu Kush range (Litvinchuk et al. 2018b).  Remarks. According to the map of Wagner et al. (2016), toads from Balkh Province should belong to B. turanensis (or B. pewzowi, B. shaartusiensis, see Ficetola and Stöck 2016;Litvinchuk et al. 2018a), from Badakhshan and Kabul Provinces to B. baturae/pseudoraddei and from Kandahar to B. zugmayeri. Although all our records represent new localities, they fall within regions or areas where these toads were previously known. Locality 5 represents an in-teresting record near Broghil Pass at an elevation of about ~3800 m. Regan (2017) identified these toads incorrectly as "Pseudepidalea viridis" and "Bufo surdus" (pp. 51-55). The genus Pseudepidalea is a less used generic name for these toads, while Bufotes surdus (Boulenger, 1891) has never been recorded in Afghanistan.

Euphlyctis cyanophlyctis (Schneider, 1799)
Distribution in Afghanistan. According to Wagner et al. (2016), this species is known from three isolated regions: first follows the valleys of Kabul and Kaitu Rivers in the east, second from Helmand River in the south and third from the area between the Khash and Farah Rud rivers in the south-west (provinces Farah, Helmand, Khost, Laghman, Nangarhar; Fig. 2).
Remarks. Distribution of this species in Afghanistan is restricted to the aforementioned river valleys or oasis and the surrounding vicinity and represents the species' northern distribution. Our record is the first for Kandahar Province and represents a new locality of the species.

Testudo horsfieldii Gray, 1844
Distribution in Afghanistan. This tortoise is probably distributed throughout Afghanistan except the central massif of Hindu Kush and its valleys (Wagner et al. 2016 and see remarks therein for additional localities with an unclear position; Fig. 2). It is currently known from provinces of Balkh, Farah, Ghazni, Herat, Jowzjan, Kabul, Kunduz and Nimroz.
Remarks. All our records represent new localities and first species observations for the respective provinces. Localities 2 and 3 represent an important range extension in the country, connecting known species distributions in Afghanistan and Pakistan (Khan 2006). Wagner et al. (2016) recognises T. h. horsfieldii in Afghanistan. However, Fritz et al. (2009) mentioned the possible occurrence of another mitochondrial clade that could represent a different subspecies. We have noticed cases in Kabul and Kandahar where tortoises were stored by local traders for the Chinese food market. Therefore, conservation programmes should be prepared for this endangered tortoise.
Remarks. Our record is the northernmost for the species in Afghanistan, only 2000 m from the border with Uzbekistan.

Trapelus agilis (Olivier, 1804)
Distribution in Afghanistan. Distribution of this species in the country is well known compared to other herpetofauna. It is known from provinces Badakhshan, Badgis, Farah, Faryab, Ghazni, Helmand, Herat, Jowzjan, Kabul, Kandahar, Takhar and Zabul (Wagner et al. 2016). The following localities, presented by Wagner et al. Remarks. All records presented here represent new locality data for the species in Afghanistan. Localities 1 and 5 represent new provincial records (Fig. 2). Regan (2017) presents this species on pp. 16-20. The individual from page 18 is presented here as Trapelus aff. megalonyx (see below).
Remarks. Both our records are new locality data. The record from Kandahar is currently the most western point of the species in the country and the record from Paktia is the first for the province. Although our records clearly correspond with the distribution range of the species in the country and individuals are very similar to T. megalonyx, photos do not allow better examination for clear species determination. Regan (2017) presents both records as T. agilis (pp. 15, 18). However, the individual from Gardez showed unusual colour patterns with an atypical violet dot on the lateral part of the body (Fig. 4H). Therefore, we present our records as "affiliated" to T. megalonyx. Remarks. This species is known from Kabul city and its vicinity. Our record clearly falls in close proximity of known localities for the species. Therefore, we did not map this record. This species is mentioned as "Unidentified Gecko" in Regan (2017;p. 29).

Cyrtopodion agamuroides (Nikolsky, 1900) complex
Distribution in Afghanistan. This species has never been recorded in the country (Wagner et al. 2016). Although Šmíd et al. (2014) mentioned its presence, they do not provide any reference or voucher specimen. On the other hand, this species is known from border areas of Iran and Pakistan (Khan 2006;Šmíd et al. 2014;Fig. 5) and its occurrence in Afghanistan was highly expected (Wagner et al. 2016).
Remarks. This is the first record of the species in Afghanistan, located approximately 250 airline km from the nearest locality in Iran (cf. Šmíd et al. 2014). Given morphological variability and ecological similarity with other members of Cyrtopodion (see Anderson 1999), this species probably forms a species complex that needs further examination.

Cyrtopodion scabrum (von Heyden, 1827)
Distribution in Afghanistan. This species is currently reported from seven provinces mainly in eastern and southern Afghanistan (Farah, Helmand, Kabul, Kandahar, Khost, Nangarhar, Zabul; Wagner et al. 2016). However, a record from Nangarhar (locality Jalalabad) is not mapped by Wagner et al. (2016), although they present its distribution in the text (see p. 487 vs. Plate 6, p. 542). One dubious record from the north-western part of the country was presented by Sindaco and Jeremčenko (2008).
Our records ( Remarks. Locality 1 in this study represents the first record for Kabul Province. Although other records represent new distribution data, they clearly fall into the known range of the species. This species is listed correctly by Regan (2017) on p. 24 but incorrectly on p. 26 (bottom photo; Lashkargāh, Helmand) as "Tenuidactylus caspius" that occurs in northern Afghanistan, but not in Helmand Province.
Remarks. This is the first record for Kabul Province and complements the species distribution in the north-eastern part of the country. This species is mentioned in Regan (2017) on p. 25.
Our records. Camp Mike Spann Chapel, Mazar-i-Sharif, Balkh (3 September 2010), one adult individual observed in a building of the camp (Fig. 6I).
Remarks. This species is already known from Mazari-Sharif and Balkh Province. Therefore, we did not map this record. The observation and species status is correctly described in Regan (2017) on p. 26 (upper photo).
Remarks. The species is mentioned by Wagner et al. (2016) from the locality "Amu- Darya swamps,nr. Darquad,N of Djangi Quala,[Takhar Prov.,400 m] (ZMK 2562)" but coordinates presented for this locality are the same as for "Amu-Darya, N of Kunduz" (see therein Appendix 1, p. 551). Therefore, the distribution point from Takhar Prov. is not mapped and presented in Fig. 5 of this study. Our record is currently the northernmost locality of the species in Afghanistan, only ca. 1500 m from the border with Tajikistan and 4000 m from the border with Uzbekistan. Although our record clearly corresponds with the distribution range of the species in the country and colouration and pattern of the species corresponds with E. nigrocellata, we present this record as "affiliated".
Remarks. All our records represent new distribution data for the species in Afghanistan. Localities 1 and 2 are new provincial records. Observations from Kandahar are incorrectly described in Regan (2017;see pp. 32, 33) as E. fasciata (juvenile presented here in Fig. 7C) and Acathodactylus micropholis (adults and juvenile in Fig. 7B, D).
Our records (Fig. 5). 1 -Lashkargāh, Helmand Remarks. Both records presented here are new locality data with original coordinates. However, they clearly correspond with the known species range in Afghanistan. Regan (2017) listed this species as "Mesalina guttulata" (pp. 30-31). This name was superseded by M. watsonana after being upgraded to full species status from a subspecies of M. guttulata. Remarks. All our data correspond with previous records of the species from close localities and particular provinces. Therefore, we did not map these records. This species is described correctly in Regan (2017) on p. 40 but incorrectly on p. 46 (both photos; Camp Dubs -Kabul, Kabul) as "Hemorrhios ravergieri" (this version uses the wrong genus name).

Ptyas mucosa (Linnaeus, 1758)
Distribution in Afghanistan. This species is known from several scattered records across the country (provinces Badghis, Faryab, Herat, Kabul, Kandahar, Logar, Nuristan,   Remarks. Wagner et al. (2016) presented the species occurrence in Kabul and its vicinity. Therefore, we did not map this record. The observation is correctly mentioned in Regan (2017) on pp. 42-43.

Spalerosophis diadema (Schlegel, 1837)
Distribution in Afghanistan. The distribution pattern of this species is divided into two main parts with desert or semi-desert habitats: the northern (provinces Herat, Faryab) and south-eastern (Kabul, Kandahar, Nangarhar). For details and other general locality data, see Wagner et al. (2016;p. 512).
Remarks. Our records correspond with the known or possible range of the species. Record 2 from Kandahar Province is the first exact record of the species from the southern part of the country and this province. The record from "Nushki to Helmand" has imprecise locality data (Wagner et al. 2016) and is most likely located in the southern part of Kandahar province. The record from Camp Dubs -Kabul was erroneously assigned to Lytorhynchus ridgewayi Boulenger, 1887 by Regan (2017;see p. 47). This species of Lytorhynchus is known in Afghanistan from only two localities of the western and south-western part of the country (provinces Herat, Kandahar).
Our records (Fig. 8). Camp Dubs -Kabul, Kabul (3 May 2011), one adult individual observed in the bushy semi-desert habitat of the camp.
Remarks. Our record is the first for Kabul province. The observation is correctly mentioned in Regan (2017;pp. 44-45, photos of the specimen are from Saudi Arabia but low quality photos are available also from Kabul, see Suppl. material 1).

Echis carinatus (Schneider, 1801)
Distribution in Afghanistan. This species has been recorded from various parts of Afghanistan (provinces Balkh, Farah, Helmand, Herat, Kandahar, Nangarhar, Nimroz;Wagner et al. 2016). These authors also mentioned a record at "Sistan [Faizabad Prov.] (ZMUC R-6838)", p. 516. This record is probably incorrect as there is no Faizabad Prov. in Afghanistan and the city Faizabad (Fayzabad) is in Badakhshan Prov. (eastern Afghanistan). Moreover, the coordinates provided by authors in the Appendix 1 are the same as for locality "Seistan [=Sistan area near Iran border]", p. 556 (western Afghanistan). Thus, potential distribution in Badakhshan needs future clarification.
Our records. Kandahar, Air Base, Kandahar (27 March 2009), adult, subadult and juvenile individuals of both sexes (Fig. 9G, H) in the desert habitat with bushes near the runway.
Remarks. This species is known from Kandahar city and its vicinity and our record corresponds with previous records presented by Wagner et al (2016). Therefore, we did not map this record.