A type catalogue of the reed frogs (Amphibia, Anura, Hyperoliidae) in the collection of the Museum für Naturkunde Berlin (ZMB) with comments on historical collectors and expeditions

We present a commented catalogue of the type specimens of the Afro-Malagasy frog family Hyperoliidae at the herpetological collection of the Museum für Naturkunde Berlin (ZMB). In current publications and databases, many names based on ZMB primary types are listed as synonyms of other species, the types often declared as lost. Consequently, the respective names are often no longer considered in current taxonomic work. We traced 146 nominal taxa of the family Hyperoliidae in the ZMB collection of which currently 130 are presented by primary types (88 holotypes, 10 lectotypes and 32 taxa based on syntype series); 50 of these taxa are currently considered as valid. Primary types of nine taxa could not be located during our inventory of the collection holdings. Seven taxa are exclusively represented by secondary types (paratypes). Many of these types comprise taxa where types have been thought to be lost. As a further service to the community, we provide important details about collectors and their travel routes, as well as respective documents stored in the collection of the Department of Historical Research at ZMB. This should make it easier to potentially compare the ZMB types in future taxonomic revisions.


Introduction
The amphibian collection of the Museum für Naturkunde Berlin (ZMB) is one of the richest in the world, comprising about 60,000 specimens of ca. 2,000 species. One of the largest and most diverse families in the collection is the African frog family Hyperoliidae with approximately 3,500 wet preserved specimens, including about 860 type specimens. These specimens have been used by various authors as a basis for 145 first descriptions, reflecting several periods of research on reed frogs by staff herpetologists and external researchers associated with the ZMB collection.
The oldest specimens from this family present in the collection are from South Africa. These vouchers, sent by G. L. E. Krebs (1792Krebs ( -1844, are two Hyperolius horstockii (ZMB 3061) from the "Cap". These specimens either were part of a shipment by Krebs containing 90 "Amphibien", in addition to other natural history specimens, and arrived at the museum on 21 June Ghana] by the zoologist and colonial administrator of the Dutch Gold Coast, H. S. Pel (1818-1876. Pel, who stayed in the Gold Coast (modern Ghana), from March to April 1841, March 1842 to December 1843, in January 1844, and from September 1847 to November 1850, collected natural history objects for the Rijksmuseum van Natuurlijke Historie in Leiden (Holthuis 1968). In March 1849, the Leiden Museum probably exchanged Pel's specimens together with other herpetological and ichthyologcial objects with the Zoologisches Museum Berlin (Lichtenstein "Eingangsjournal ZMB"). The first mention of Hyperolius species from the ZMB was in the catalogue published by Lichtenstein and Martens (1856: 36) "Nomenclator Reptilium et Amphibium Musei Zoologici Berolinensis".
A decade later, at the beginning of 20 th century, large scale research expeditions such as the first and second "Deutsche-Zentral-Afrika-Expedition" [German-Central-African-Expedition] from 1907-08 and 1909-10, provided rich new material. Together with specimens obtained from the German colony Cameroon, further remarkable collections of amphibians and reptiles, including considerable numbers of reed frogs, were sent to ZMB, and were partly described by Friedrich Erich Gottlieb (called Fritz) Nieden (1883Nieden ( -1942 (Nieden 1910a(Nieden , 1913; see also Günther and Bischoff 2018).
In December 1921, Christoph Gustav Ernst Ahl (1898-1945 started to work as a voluntary scientific assistant in the ichthyological and herpetological department of ZMB. After Tornier's retirement, Ahl became an official assistant from February 1923to September 1927. From October 1927until his dismissal in March 1941, he was employed as senior scientific assistant in the herpetological department of ZMB. During his career, Ahl published a total of 302 new descriptions of amphibians and reptiles (Paepke 2013). Over a period of seven years, he described 106 nominal taxa from the family Hyperoliidae (Ahl 1924(Ahl , 1930a(Ahl -d, 1931a, and worked on this family monographically, published as part 55 of the "Tierreich" [animal kingdom] series (Ahl 1931b). However, Ahl's work is overshadowed by a large number of scientifically flawed descriptions. A selection of Ahl's Hyperolius types were sent in exchange to the Museum of Comparative Zoology at Harvard University (MCZ) in 1932, and this material allowed Barbour and Loveridge (1947) and Loveridge (1957) to synonymize 50 of Ahl's species. Arthur Loveridge (1957: 157, 324) called Ahl a "reckless describer" and complained about his insufficient knowledge of intra-specific variability within the Hyperoliidae as well as his inadequate comparisons with already described species. Ahl's working methods and the quality of his descriptions have subsequently been commented on, e.g. by Glaw and Vences (1992), Paepke (1995Paepke ( , 2013, Rieck (2001), and Adler (2007). In addition to Ahl's questionable scientific approach, many of the (type) specimens mentioned by him (Ahl 1931a, b) were not inventoried and placed in the systematic collection during his tenure; instead, these were often placed on the shelves of undetermined material, sometimes with handwritten notes. In the 1950s Heinz Wermuth , then curator of herpetology at ZMB, at the request of Raymond Ferdinand Louis-Philippe Laurent (1917Laurent ( -2005, endeavored to locate Ahl's type material. In September 1958 and January 1959, a total of 244 Hyperolius specimens, including the locatable types, were sent on loan to Laurent at the "Université Officielle du Congo-Belge, Elisabethville" [Lubumbashi, Democratic Republic of the Congo]. Based on this material Laurent (1961) published a paper on the Hyperolius and Afrixalus in the ZMB. He reviewed some of Ahl's taxa and mentioned for the first time respective inventory numbers. In the introduction of this paper he suggested that it would be desirable to declare all descriptions based on the activ-ity of this individual zoologist [Ahl] as nomina confusa (see also comments in Liedtke et al. 2014: 254).
Several 'Ahl species' were synonymized by Loveridge (1942Loveridge ( , 1957; Laurent (1943Laurent ( , 1958Laurent ( , 1961 and Barbour and Loveridge (1947). Other herpetologists studied Berlin hyperoliid specimens in the second half of the 20 th century. Among them were Arne Schiøtz (1932Schiøtz ( -2019 from the Zoological Museum Copenhagen, who got material on loan in 1960and 1961, and visited the ZMB in 1968 in order to study the East African frog collection, and Jean-Luc Perret from the Natural History Museum Geneva, who borrowed a number of Ahl types in 1962, and visited the ZMB collection in 1974. As a result of their studies, further 'Ahl species' were synonymized (see e.g. Schiøtz 1967Schiøtz , 1975. Only 24% of the reed frogs described by Ahl are still considered valid (see below ;Paepke 2013;Frost 2021).
After 2000, ZMB received new hyperoliid vouchers from West and Central Africa, as well as from Mozambique, mainly through the collections of the working group of Mark-Oliver Rödel, curator of herpetology at ZMB since 2007. J. M. Dehling (University Koblenz-Landau) provided new vouchers from Rwanda, V. Mercurio (Berlin) collected in Malawi, and A. Channing (University of the Western Cape) sent vouchers from South and East Africa. The study of these new collections and the re-evaluation of historic specimens resulted in a number of revisions and new descriptions (see e.g. Rödel et al. 2003Rödel et al. , 2009Rödel et al. , 2010Dehling 2012;Channing et al. 2013;Frétey et al. 2014;Liedtke et al. 2014). These and other studies, often based on 'new' molecular technologies, also revealed that the diversity within the Hyperoliidae might be much higher than previously assumed and comprise a large number of cryptic taxa (i.e. Channing et al. 2013;Portik et al. 2019). Thus it is likely that some of the ZMB 'synonyms' actually refer to valid species (see Rödel et al. 2010). Current researchers describing new African reed frogs tend to ignore the availability of these names because these are listed as either synonyms or lost by Frost (2021). However, the collection of hyperoliid frogs in ZMB was never fully reviewed and the status of several taxa and the presence of many type specimens remained unevaluated. For instance, no fewer than 17 nominal taxa have been reported lost or not traced by Frost (1985) or listed as originally present in ZMB but without inventory number (Frost 2021).
We present a list of existing and so far unlocated type specimens of 146 nominal taxa of the family Hyperoliidae from the ZMB collection of which 130 are primary types (88 holotypes, 10 lectotypes and 32 taxa based on series of syntypes), 50 of which are currently considered valid. Primary types of nine taxa could not be located. Seven taxa are represented by secondary types (paratypes) only. This summary is not intended to resolve pressing taxonomic issues that in many cases will require dedicated research using molecular genetic approaches (see Scherz et al. 2020 for a promising example). In some cases, however, we added observations of taxonomic value (e.g. measures and character descriptions), clarified priority of names, and identified and corrected type localities when possible. Our aim is to present what is available for study to aid future research that describes and names new taxa in the Hyperoliidae. This contribution is the third catalogue on the amphibian types held by ZMB. Previous catalogues include the caecilians and salamanders (Bauer et al. 1993), as well as the hemisotid, microhylid, myobatrachid, pelobatid and pipid frogs (Bauer et al. 1996).

Methods
From 2017 to 2020, all specimens from the family Hyperoliidae in the collection of the Museum für Naturkunde Berlin (ZMB; in some publications the Museum für Naturkunde is also abbreviated with MfN or other acronyms (see Sabaj 2020); to avoid confusion, we apply the traditional use of ZMB for the herpetological collection) were systematically digitally registered, including all specimens not previously inventoried. Lots-jars with specimens all carrying the same accession number-were individualized and each specimen was assigned an individual number. Details on the original field labels were compared to catalogue entries, information from the accession catalogues and the data were completed where necessary. ZMB numbers always refer to the final inventory catalogue numbers. In addition, we sometimes mention 'accession numbers'. These are separate catalogues used in the herpetological department from ca. 1856 to April 1940, to record the accession of new material, often registered in lots with "C-Catalogue" numbers ( Fig. 3). Only after having been assigned a 'ZMB number' are vouchers finally inventoried. The terms 'Register Catalogue' or 'ZMB Register' refer either to the accession catalogues of the Zoological Library or to Lichtenstein's 'Eingangsjournal' for the entire Zoological Museum, the latter archived in the Department of Historical Research at the Museum für Naturkunde (see Unpublished Sources). Our digital accessioning of the specimens, in connection with information on collectors, localities and collection periods, made it possible to search for previously unlocated type specimens of the species described by Ahl (1931a). For primary type specimens, the type localities and the collectors were identified whenever possible. Secondary types from the Berlin collection are likewise listed with locality and collector. If not stated otherwise, we follow Frost (2021) regarding the currently valid names. Concerning type localities, we always provide first the original spelling (in quotation marks) followed by the currently applied name of the locality in brackets, as well as further geographic data, such as province and country. We list all types in alphabetic order, using the original name in the description. The present taxonomic status and generic association is given under 'present name' in each species account. Remarks in the individual species accounts contain information on illustrations of the type material, the activities of the collectors, and the collection periods as well as information taken from secondary literature on additional type material that is not housed in the ZMB collection. We omit providing information on the history of the synonymy of the respective taxa. This is provided by Frost (2021) and can be consulted at: https:// amphibiansoftheworld.amnh.org/index.php. It is important to note that we herein do not take any taxonomic decisions; we present the available type specimens and their current taxonomic name. For valid names we refer, with few exceptions, to Frost (2021). Until the complex taxonomy of the H. marmoratus species group is solved and allows a more accurate assignment, we tentatively follow Marques et al. (2018) and list Angolan taxa previously considered as synonyms of H. marmoratus Rapp, 1842or H. parallelus Günther, 1858 as belonging to Hyperolius angolensis Steindachner, 1867. Ceríaco et al. (2020 confirm assignment of Angolan frogs of the H. marmoratus group to H. angolensis. In contrast Frost (2021) and Baptista et al. (2019) regard H. angolensis as a junior synonym of H. parallelus Günther, 1858. An even more conservative approach has recently been applied by Channing and Rödel (2019), treating almost all populations of the Hyperolius viridiflavus/marmoratus-complex as H. viridiflavus.
Recently, Dubois et al. (2021) suggested 'new concepts and methods for phylogenetic taxonomy and nomenclature', using Lissamphibia as a template. To avoid causing further confusion in an already complex and taxonomically confusing group of tropical tree frogs, we -without assessing the new system -refrain herein from following Dubois et al. (2021). The future will show if researchers will accept and apply this new concept.
In the chapter "Specimens erroneously marked as types in ZMB inventory catalogues", we mention names which are labeled as types in the inventory catalogues of the Herpetological Department of ZMB. To our knowledge, these names were never published by the authors to which these names are attributed, nor by anyone else. These names are thus placed in quotation marks and not italicized to indicate that they are not used as valid. Since these names were sporadically used (e.g. Schiøtz 1975) we list them here and clarify the identity of the specimens if possible.
It is important here to provide commentary on the publication history of Ahl's (1931a, b) works on hyperoliid frogs and how that relates to priority of the species names published therein. In 1931, Ahl published two monographs: the paper "Zur Systematik der afrikanischen Baumfroschgattung Hyperolius [Towards the systematics of the African tree frog genus Hyperolius] (Ahl 1931a)" as well as Lieferung [issue] 55 "Anura III" from the 'animal kingdom' series ["Das Tierreich"] (Ahl 1931b). The later work covered all known frog species of what today is accepted to be the family Hyperoliidae. However, it has never been fully clarified which of these monographs was published first, and thus which of these two publications serves as the original publication for making available the Hyperolius species names described by Ahl in 1931. We thus researched respective entries in the zoological main library of ZMB and in the reprint collection of the herpetological department. On the front covers of both publications we found the following notes on publications dates: volume 17, issue 1 of the "Mitteilungen aus dem Zoologischen Museum in Berlin" (Ahl 1931a) was published on 1 April 1931; Lieferung (volume) 55 of Anura III of the series "Das Tierreich" (Ahl 1931b) was published in March (März) 1931 (Fig. 4). Later, these publication dates were often overlooked or ignored. Hence numerous authors gave priority to the names published in Ahl's Lieferung 55 of the "Das Tierreich" (e.g. Laurent 1941Laurent , 1958Loveridge 1942Loveridge , 1953Loveridge , 1957Barbour and Loveridge 1946;Manaças 1949, Perret andMertens 1957;Schiøtz 1975;Frost 1985;Channing and Howell 2006;Pickersgill 2007a;Seniagbeto et al. 2007;Mercurio 2011;Dehling 2012;Paepke 2013;Frétey et al. 2014;Liedtke et al. 2014;Marques et al. 2018), others however, regarded the names published in the "Mitteilungen vol. 17(1)" as having priority (e.g. Loveridge 1936a, b;Laurent 1943;Schiøtz 1967;Perret 1976b;Pakenham 1983;Rödel 1996;Rödel et al. 2010;Amiet 2012). Frost (2021) states "The description in Ahl, 1931[a], Mitt. Zool. Mus. Berlin, 17 […], appeared a few weeks later according to unpublished notes by A. Loveridge (fide R. Laurent)." These personal notes and comments by Arthur Loveridge and Raymond Laurent correspond with the printed publication dates on the front covers of the two publications, as has been already commented on by Barbour and Loveridge (1946, p. 126).
However, it needs to be emphasized that, without doubt, Ahl intended to publish the paper "Zur Systematik […]" (Ahl 1931a) ahead of the "Anura III" (Ahl 1931b). For instance: i) in the preface of his paper" Zur Systematik […]" he refers to the ‚soon to be published monograph within the ‚animal kingdom series' ["demnächst erscheinende Monographie im "Das Tierreich""]; ii) in the paper he added "spec. nov." to the new names, in the species accounts of "Das Tierreich" this is not added, and iii) lastly comments concerning the distribution of species are generalized in "Das Tierreich" and data concerning type material and collectors are lacking completely (but are provided in "Zur Systematik […]").
To finally clarify the history of both publications (Ahl 1931a, b), we checked the original prints and Eingangsregister (accession catalogues) in the department of herpetology and the zoological library at ZMB. Based on the receipt stamps on the original prints (journal issue and reprints) as well as the notes in the Eingangsregister [entry register] of the zoological library, it is obvious that volume 17, issue 1 of the "Mitteilungen aus dem Zoologischen Museum in Berlin" was received on 11 April 1931. In contrast Lieferung 55, Anura III from "Das Tierreich" was only received a month later on 11 Mai 1931. Article 21.4 of the `Code' (ICZN 1999) clarifies that "If the date of publication specified in a work is found to be incorrect, the earliest day on which the work is demonstrated to be in existence as a published work is to be adopted". Furthermore, Recommendation 21D states that "A librarian should not remove, or allow to be removed by a binder, the cover or pages bearing information relevant to the date of publication, the contents of the work or its parts, or the day or dates of receipt by the library." This makes the receipt dates from the ZMB library relevant in this context. We thus follow Recommendation 21F of the `Code' and correct the publication dates which gives the names published by Ahl (1931a)
Afrixalus brevipalmatus see Hyperolius brevipalmatus. it does not specify a specific geographic region but refers generally to regions being away from the coast or provincial towns] and to Yaoundé and founded his own plantation in the border area between German-Cameroon and Spanish-Guinea.  Ahl (1931b: 282, fig. 153). For the location of "Konde-Land" we refer to Fülleborn (1906: 268 ff.), who describes it as a small area at the northern tip of Lake Malawi as follows: limited in the east by Lake Malawi, in the northeast by the slopes of the Livingstone Mountains [Kipengere Range], in the southeast by the Untali and Malila Mountains, in the north by Rungwe Volcano and in the south by the lower reaches of the Ssongwe (Songwe River).
In 1896, the German physician, doctor of tropical medicine, and explorer Fülleborn joined the colonial "Schutztruppe" in German East Africa, where he was active as a government physician until 1901. From April 1897 to the beginning of 1898, he participated in the military campaigns against the Wangoni and Wahehe of the "Ungoni", "Uhehe" and "Ubena" regions, in present day southern Tanzania. From 1898 to 1899 Fülleborn was stationed in Langenburg [Lumbira, Mbeya Region, Tanzania] in the north of Lake Malawi and undertook numerous excursions in the surrounding area, which took him to the southern end of Lake Malawi, through the "Schire-Hochländer" [Shire Highlands, southern Malawi], and on the Shire and Zambezi River to Quelimane and afterwards to the Island of Mozambique. In 1899 he was commissioned to research the "German-Nyassa" region from a zoological and ethnological-anthropological point of view. Together with W. Goetze he participated in the "Nyassa-See-und Kinga-Gebirgs-Expedition" (Engler 1902;Fülleborn 1906; see also comments on Hyperolius goetzei). During this time Fülleborn also surveyed Lake Malawi and the lakes in northern Nyasaland (Rukwa, Chungruru, Itende) and collected a considerable number of mammals, about 800 birds, more than 1000 fishes, amphibians and reptiles, thousands of insects and other invertebrates, and particularly plankton (Fülleborn 1900a, b;Paepke and Seegers 1995). The majority of these collections were donated to ZMB. Fülleborn's extensive herpetological collections were partly studied in the first third of the 20 th century by former curators of herpetology at ZMB (e.g. Ahl 1929Ahl , 1931aTornier 1900Tornier , 1902Tornier , 1905 Remarks. Perret (1966: 408) considered the type material of H. acutirostris lost and designated MHNG 965.12 as neotype. Bauer et al. (1995: 43) could only locate one of the two syntypes. The type locality was corrected to "Douala" by Frétey et al. (2014); for further information see also remarks on Hyperolius guttatus.
The German explorer, zoologist and anatomist Buchholz went to Equatorial Africa from 1872 to 1875. He was accompanied by the Berlin ornithologist Georg Anton Eugen Reichenow and Reichenow's friend, fellow student and zoologist Wilhelm Lühder. On 1 June 1872 they set off from Bremerhaven to "Akkrá on the Gold Coast" [Accra, Ghana], which they reached on 29 July 1872. The first collecting tours took place in the surroundings of Accra and Aburi (29 July to 16 October 1872). On 16 October they left Accra for "Camaroons" [today part of present day Douala city] where they stayed until 2 November. Then they travelled to Bimbia, Victoria and Bonjonjo (2 November 1872 to 9 December 1873). On 12 March 1873 Lühder died of malaria in 'Camaroons'. Reichenow, also suffering from malaria, returned via Gabon to Germany in April 1873.
Buchholz was on his own from then on. He travelled between Victoria and `Camaroons' with intermediate stops on Fernando Pó to get his collections to Camaroons in early December.
Thereafter he went to Abo (9 December 1873 to 24 March 1874) and from Mungo via Balong he returned again to 'Camaroons' (5 April to 11 August 1874). He left `Camaroons' for a stay in Gabon where he also explored the Rembo River (12 August to 9 November 1874). After his return to the Gabon coast he again explored the area around Mungo and Jenssoki (9 November 1874 to 11 January 1875), and again visited Fernando Pó, the Gabon coast and the Ogowe (or Ogooué) River (11 January to 31 August 1875). On 3 September he started from Gabon on his way back to Greifswald where he arrived during the beginning of November 1875 (Reichenow 1874;Heinersdorff 1880;Weidmann 1894;Stresemann 1943). Buchholz' collections went to the zoological museums in Greifswald and Berlin and his herpetological material has been described by Wilhelm C. H. Peters (Peters 1875(Peters , 1876.  fig. 171). Originally eight specimens according to the original publication. Lectotype designation by Perret (1962: 244, fig. 2) who provided a photograph of the lectotype. Another paralectotype MCZ A-17626 was sent in exchange from ZMB in 1932 (Barbour and Loveridge 1946: 126). Adametz was a first lieutenant in the German `Schutztruppe' for Cameroon and head of the colonial station in Bamenda. He was involved in surveying the Hinterland of the Kamerun-Nordbahn in the Bamenda region. In summer 1912, he also took part in an operation against the Baminge (Bamije-Expedition) at the eastern frontier of the present day Manyu Division, Southwest Region, Cameroon (Nkwi 1989;Hoffmann 2007;Hafeneder 2008 (Schubotz 1909(Schubotz , 1912Bamps 1975). The extensive zoological-botanical collections made during this expedition, including nearly 3000 vertebrates, were deposited at the ZMB and the Botanical Museum in Berlin. Most of the herpetological results of the expedition were published by Nieden (1913, Amphibia) and Sternfeld (1913, Reptilia).

Hyperolius ademetzi
From 1909 to 1910 a second "Deutsche Zentral-Afrika-Expedition", also under the leadership of Adolf Remarks. The "Africanische Gesellschaft", or formally "Deutsche Gesellschaft zur Erforschung Aequatorial-Africas", sponsored the "Loango-Expedition" from 1873-1876 under the leadership of the German geographer and explorer Richard Paul Wilhelm Güssfeldt. The expedition had the task to establish a station at the Loango coast (at Chinchoxo), which was to serve as a depot for the material collected during the expedition. Geographic-topographical explorations into the interior of the African continent were also intended to be carried out. With an interdisciplinary research team, comprising the medical officer and zoologist Julius Falkenstein, the geographer Eduard Pechuël-Loesche, the geodesist von Görschen, Reserve Lieutenant Hans von Hattorf, the mechanic Otto Lindner, the botanist Herman Soyaux, and the topographer Major Alexander von Mechow, Güssfeld travelled for two years, starting in July 1873. They mainly followed the coastal area of Cabinda, on the Kouilo river, the Chiluango river, and on the lower course of the Nyanga river. From March 1874, they turned to Luanda, on the Cuango to Dondo and to the rapids of Cambambe, as well as to Quicombo and Novo Redondo (Güssfeldt et al. 1879, Weidmann 1894, Heintze 2007, Marques et al. 2018. The amphibians and reptiles collected during these trips were sent to ZMB and described by Peters (1877a, b).  fig. 189). There remains confusion about the collector and, related to this, the likely place of collection. A man by the name of Ernst Heinrich Georg Ule collected in Brazil and donated two frogs to the herpetological collection, one with the accession catalogue number C-581 (from November/ December 1912) without further data, and a second one (C-145) on 21 May 1904, collected on the Upper Amazon. The frog thus might actually be a South American tree frog and not a hyperoliid. However, another person named Dr. Ferdinand Uhl was a member of the "Deutsche Schutztruppe" in East Africa who collected the holotype of Hyperolius guttolineatus Ahl, 1931 (see below, unlo-cated type specimens). Lastly, a person with the surname Uhle collected in Sumatra, Bolivia and Argentina. Thus, neither the identity of the frog, nor its geographic origin and collector can be determined with certainty.  (Hösemann 1897, Hafeneder 2010). Among others, he participated in the "German-French Border Expedition" (October 1901to December 1902 to define the southern border of Cameroon, during which he mapped the area from the camp Nyengwe, south of Kampo, to the Ngoko station in the Sanga Ngoko area (Danckelmann 1901;Fitzner 1901 Ahl, 1931a. Remarks. Drawing in Ahl (1931b: 291, fig. 167). According to Ahl (1831a: 35) the original series consisted of four specimens, i.e. one collected on 26 July 1894 and three on 07 May 1893. However, the latter year given by Ahl is most probably a typographical error. According to Baumann's preserved original field label, the date of collection was the "7. Mai 1894". Another paratype MCZ A-17627 from "Misahöhe, Togo", coll. Baumann on 07 May 1894, was sent in exchange from ZMB in 1932 (Barbour and Loveridge 1946: 127). The latter specimen was erroneously regarded as a holotype by Seniagbeto et al. (2007: 77 Remarks. Drawing in Ahl (1931b: 322, fig. 196). Another paratype MCZ A-17628 from "Konde-Nika", coll. Fülleborn was sent in exchange from ZMB to MCZ in 1932 (Barbour and Loveridge 1946: 127).  Present name. Afrixalus fornasini (Bianconi, 1849). Remarks. Depicted in Peters (1882b, pl. 24, fig. 2 and pl. 26, fig. 6, sternum). Peters (1882b: 161) specified that he found this species in March 1846, often in grass and on bushes in the Prazo [estate] Boror northwest of Quellimane. Poynton and Broadley (1987: 192) Ahl (1931b: 316, fig. 190). Another paratype, MCZ A-17629 from "Eldama River Station", coll. Grote was sent in exchange from ZMB in 1932 (Barbour and Loveridge 1946: 127).

Hyperolius bivittatus
In October 1905 and November 1907, Tiesler sent two shipments, including nearly 300 amphibians and reptiles, to ZMB. This material was collected between November 1904 and January 1906 in Portuguese East Africa [Mozambique] and described by Nieden (1915). The vouchers of this collection originated from the following localities: Cabayra, Chifumbazi, Chinta, Costa, Lukunga, Marazi, Missala, Tschimbo, Tschinoupe and from the Zambezi River without any exact locality data. However, Ahl's description is clearly based on a single specimen "1 Stück [piece]" from "Sangmelina" purchased from "Rolle". Furthermore, the collection data of ZMB 20132 from "Bipindi" [Bipindi village, Océan Department, South Province, Cameroon], coll. Georg Au-gust Zenker, do not match the information provided in the original description.

Hyperolius brevipalmatus
Rolle was a well-known dealer of zoological and enthnological objects. He maintained a worldwide network of collectors and suppliers and acquired several important collections. From 1889 onwards, he supplied private collectors as well as important European museums with zoological objects from Berlin. In later years, he traded objects under the name of the natural history institute "Kosmos".

Hyperolius callichromus
The Austrian hunter and Africa explorer Grauer undertook several expeditions to Eastern Africa, e.g. to British East Africa [Uganda] (February to May 1904 andSeptember to November 1905) and to Tanganyika in 1907, where he met the first "Deutsche Zentral-Afrika-Expedition" at Lake Kivu in August. Upon this meeting he handed the zoological material he had collected in the "Zwischenseengebiet" [Region between Lake Victoria, Lake Kivu and Lake Malawi, Tanzania] for ZMB and the Walter Rothschild Zoological Museum (now the Natural History Museum at Tring), to the German expedition. Grauer then turned south, travelled along the west bank of Lake Tanganyika and returned to Europe in early 1909 (Schubotz 1909(Schubotz , 1912ÖAW 1959;Riedl-Dorn 2001). In November 1909, he returned to Africa, on behalf of the Natural History Museum Vienna (NMW) and travelled to Lake Victoria and Lake Malawi. The herpetological collections of Grauer's last expedition were partly described by Steindachner (1911) and Werner (1924). We refer also to Gemel et al. (2019) for information about the type material collected by Grauer and deposited in the NMW collection.  fig. 159). The German medical doctor von Raven, who specialized in bacteriology and tropical medicine, accompanied the first "Deutsche Zentral-Afrika-Expedition" under the leadership of Adolf Friedrich, Duke of Mecklenburg from 1907(Schubotz 1909 (Günther 1906 fig. 280). Another paratype, MCZ A-17633 from "Westliches Russisi-Ufer und Nordwest-Ufer des Tanganyika", coll. Grauer, was sent to MCZ in exchange in 1932 (Barbour and Loveridge 1946: 126 Remarks. Lectotype designation by Perret (1962). Another paralectotype MCZ A-17632 from "Longa", coll. Brühl and Gleim, was sent in exchange from ZMB in 1932 (Barbour and Loveridge 1946: 127). Drawing in Ahl (1931b: 352, fig. 227). Ahl (1931a: 80) stated that seven specimens were collected by "Brühl and Gleim". However, these two people were not active in Angola at the same time (see below). It is possible that specimens of both collectors were stored together. Thus, it is no longer possible to assign the specimens to one collector.

Hyperolius castaneus
Gleim was Deputy Governor of the German Colony of Togo from 1896 to 1898. From 1899, he was sent to São Paolo de Loanda by the "Kolonialabteilung des Auswärtigen Amtes", where he served as the first professional consul for Angola and French Congo. From 1910 to 1911 he was Governor of Cameroon (Schnee 1920a Remarks. Ahl (1931a: 109) incorrectly names "Teisler" as collector. The exact position of the type locality remain dubious. There is also a Cabaia in the Zambézia province, district of Namacurra, area of Macuze, but it is unclear if this locality was intended.  Ahl, 1931. Remarks. Drawing in Ahl (1931b: 364, fig. 239). According to Ahl (1931a: 90) the original series consists of seven specimens from "Rugegewald", including the "Type" and from "westlich des Albert-Edward-Sees's" [west of Lake Edward, Democratic Republic of the Congo], all coll. Grauer. A paratype MCZ A-17634 from Lake Edward, coll. Grauer was sent in exchange to MCZ in 1932 (Barbour andLoveridge 1946: 128). The remaining five paratypes could not be located. Liedtke et al. (2014) regarded ZMB 36089 as lectotype and restricted the type locality to "Nyungwe Forest (most likely Rwasenkoko [Uwasenkoko])".

Hyperolius flavoguttatus
The German zoologist, cartographer, explorer and colonial official Stuhlmann spent a total of 14 years in East Africa. With the financial support of the Akademie der Wissenschaften [Academy of Sciences] zu Berlin, he investigated the coastal regions of Zanzibar and the adjacent mainland including "Usegúa" and "Ungúu" in present-day Tanzania in the summer of 1888; then, until mid-1889, the area of the Zambezi estuary around Quelimane in Mozambique. From April 1890 to 1892, Stuhlmann participated as a scientist, together with Lieutenant Wilhelm Langheld, on the expedition of Mehmed Emin Pasha [also known as Eduard Karl Oskar Theodor Schnitzer] to the German East African colonial area. The expedition led them from Bagamoyo (26 April 1890) via Tabora (29 July) to Bukoba on Lake Victoria (November 1890). From here, Stuhlmann undertook a trip on Lake Victoria to Murchison Bay in Uganda (December 1890) and reached Mengo Mountain (26 to 29 December) via Manjongo [Rubaga Division, Kampala District, Central Region, Uganda]. After his return to Bukoba, he set off (12 February 1891) towards the west in the Karagwe Region, and after crossing the Kagera River (06 April 1891), the expedition reached the southwestern tip of Lake Edward in early May 1891. The expedition turned west of Lake Edward another 250 km to the north, but was terminated in mid-September 1891 due to insurmountable difficulties. With a group of 27 askaris (local soldiers serving in European colonial armies) and 100 porters Stuhlmann went back to Bukoba, where he arrived on 17 March 1891. Emin Pasha, in contrast decided to stay behind with sick expedition members, turned southwest towards the Congo River and was murdered by Arab slave traders 80 km from this destination at Kinene on Mwiko River on 20 October 1892. Stuhlmann's herpetological collections from these expeditions were sent to Johann Georg Pfeffer at the Zoologische Museum Hamburg, who published the first results (Pfeffer 1889(Pfeffer , 1893 fig. 4). Bauer et al. (1995: 44) regarded two specimens, i.e. ZMB 6631 and ZMB 6632 as syntypes of H. flavoviridis. However, Peters (1854: 628) mentioned only material from "Boror" in his original description and he specified later (1882b: 164) that he got only one male from that locality, which corresponds to ZMB 6631. Although ZMB 6632 from "Halbinsel Cabaceira" [Peninsula Cabaceira, Mossuril District, Nampula Province, Mozambique], collected in June 1843, is marked by Peters' hand as type of H. flavoviridis in the ZMB inventory catalogues, the stated locality does not correspond with the type locality. Another two specimens from "Tette" donated from ZMB to the collection in Leiden (RMNH RENA-1780 and 1785) have been regarded as possible syntypes of H. flavoviridis (Bauer et al. 1995: 44, Gassó Miracle et al. 2007. Both specimens can be excluded as types of H. flavoviridis because of the locality information, being different from the type locality. Likewise they cannot be the types of H. tettensis because of the single female type specimen mentioned by Peters (1882b: 164) is ZMB 4812 (see below). , 2013: 20, fig. 4D, fig. 6 (Fülleborn 1900a, b) (Fig. 7). Present name. Hyperolius marmoratus Rapp, 1842.

Hyperolius fusciventris
The name Rappia granulata Tornier (1896: 151) is categorized as nomen inquirendum, "Name(s) unassigned to a living or extinct population" by Frost (2021) who placed the type locality "Tette" mistakenly in Tanzania. Tornier (1896) attributed the authorship of this name to Peters and mentioned the type specimens by number (ZMB 4811). However, Peters never described a reed frog with the specific epithet "granulata". Tornier's name granulata does not meet the requirements of Art. 33.2 of the 'Code' (ICZN 1999) for an "emendation". We therefore consider granulata Tornier, 1896 as an incorrect subsequent spelling of the specific epithet granulosus Peters, 1867. Remarks. Lectotype by subsequent designation of Laurent (1961: 73). Frétey et al. (2014) corrected the type locality to "Douala" based on an account and a map of Buchholz' Central African travels provided by Heinersdorff (1880). According to the latter, Buchholz visited "Cameroons" between October 1872 and August 1874. According to Frétey et al. (2014) the collection in RMNH holds four additional paralectotypes (RMNH RENA 1788 A-D) from "Boutry", coll. Pel (not listed by Gasso Miracle et al. 2007). Drawing in Ahl (1931b: 354, fig. 229) figuring paralectotype ZMB 4489. For the origin, history and status of Hyperolius guttatus and drawings, photographs and redescriptions of the ZMB type specimens we refer to the revision by Frétey et al. (2014).
The Berlin ornithologist Reichenow travelled together with Lühder and Buchholz from spring 1872 on a one year collecting trip to "Akkrá" on the Gold Coast [Accra, Ghana] and the region around "Camaroons" [Douala Region, Cameroon] (Reichenow 1874;Heinersdorf 1880;Weidmann 1894;Stresemann 1943; see also remarks on Hyperolius acutirostris). Reichenow was assistant in the fish and reptile department in ZMB until Wilhelm Peters' death. From 1883 he worked as an administrator and assistant in the mammal and reptile department. In 1888 he became curator for the reptile, bird and mammal exhibition in the new ZMB building on the Ivalidenstraße. After the retirement of his father-in-law Jean Louis Bennoit Cabanis in 1892, and after almost twenty years conducting various activities at ZMB, Reichenow took over as the curator of the ornithological collection (Stresemann 1943 -XI/1907;ZMB 22321, coll. Lieutenant Naumann, 1911; all specimens from "Bamenda". Present name. Hyperolius riggenbachi (Nieden, 1910).

Hyperolius hildebrandti
Remarks. Depicted in Ahl (1931b: 334, fig. 207), copied from Peters (1875, pl . 2, fig. 3). The same specimen that is the holotype of H. hildebrandti is also the lectotype of H. guttatus Peters, 1875. For the origin, history and status as well as type localities, drawings, photographs and redescriptions of the ZMB types of H. guttatus and H. hildebrandti, we refer to the revision by Frétey et al. (2014).  Ahl, 1931a. Remarks. Depicted in Ahl (1931b: 374, fig. 249). The type locality indicated by Ahl (1931a: 102) is misleading and composed of two different regions which are far apart. Houy was a member of the "Lagone-Pama-Expedition 1912-13", which he accompanied as government doctor and zoologist to "Neu-Kamerun", and several natural history objects from this expedition were sent by him to ZMB. Together with the topographer and First Lieutenant Otto Tiller, he also accompanied the "Expedition ins Zwischenseengebiet in Ostafrika" [region between Lake Kivu and Lake Victoria] in 1911, the expedition directed by the colonial geographer Hans Heinrich Josef Meyer. According to the original label, the holotype of H. houyi was collected on 29 November 1911, at the end of Meyer's expedition to East Afrika (see also Urban 1917). On the basis of the map showing the expedition route (Meyer 1913), the corrected type locality for H. houyi has to be "SW-Ussagara" [southern Kilosa District, Morogoro Region, Tanzania]. Remarks. Drawing in Ahl (1931b: 284, fig. 157), copied from Bocage (1868: 844, fig. 2). The Berlin syntype was sent in 1869 in exchange from Lisbon by Bocage and was mentioned and depicted by Tornier (1896: 143, pl. 4, fig. 48). The syntypes MBL T. 21-164, 27-167 from "Benguella", coll. Anchieta and "St. Salvador du Congo" coll. António José de Sousa Barroso were destroyed by a fire in the Museu Bocage on 18 March 1978 (Marques et al. 2018: 90). Perret (1976a: 28) corrected the type locality to "São Salvador do Congo, Angola, and Novo Redondo, Angola". The Berlin syntype is not mentioned by Marques et al. (2018), but probably is the only remaining syntype.  (Duméril & Bibron, 1841).

Hyperolius leptosomus
The zoologist and anatomist Böhm, together with the explorer Paul Reichard, travelled on behalf of the "Africanische Gesellschaft" from Zanzibar via Bagamojo [27 July 1880] to Tabora, which they reached two-and-a half months later. From here they turned to Kakoma [southeast of Tabora, Tabora Division, Tanzania], where they stayed for over a year. Then they continued to Jagonda [just northeast of Kakoma]. From Jagonda, Böhm and the topographer Emil Kaiser went on a journey to Lake Tanganyika, lasting several months. They reached Karema on the western shore of the lake [Mpanda District, Katavi Region, Tanzania] and returned to Jagonda on December 23, 1881. In March 1882 they travelled along the Wala River. Dr. Kaiser died during an expedition to Lake Rukwa near Upia on 27 October 1882. Towards the end of December 1882, Böhm and Reichard left Jadonda for Karema, crossed Lake Tanganyika to Mpala (at the mouth of the Lufuku River, Tanganyika Province, Democratic Republic of the Congo) and reached the "Marunga Land" in July 1883. From here they turned southwest and discovered Lake Upemba in the Urua region [Upemba, Bukama Region, Haut-Lomami Provinz, Democratic Republic of the Congo]. On 27 March 1884 Böhm died in southern Urua, south of the Lake Upemba (Schalow 1888, Weidmann 1894. Based on Böhm's itinerary, it is clear that he found the holotype of H. marungaensis in the Marunga Highlands, where he collected extensively in summer 1883 (Schalow 1886(Schalow , 1888. Therefore, we correct the type locality to "northern Marunga or Marungu Region southwest of Lake Tanganyika, Kalemie Territory, Tanganyika Province, Democratic Republic of the Congo".  Frétey et al. (2014).
Because of a reading error, assuming ZMB 10100 instead of ZMB 10101, a new label was written for this collection jar in the 1920s, for which erroneously the information of ZMB 10100 was adopted, viz. "Nossi-Bé" and "Museum Senckenberg". This transmission error and the specimens became the basis for Ahl's (1930) new description of H. nossibeensis. In 1992, Frank Glaw (ZSM) located the syntypes of H. nossibeensis in the ZMB collection. The jar with the label from the 1920s mentioned Mantidactylus granulatus from Nosy Be, ZMB 10100. Glaw and Vences (1993: 216) discussed the status and identity of H. nossibeensis, synonymized it with Hyperolius marmoratus and corrected the terra typica to "das Äthiopische Afrika" [Ethiopian Africa]. Subsequently the three syntypes were re-inventoried as ZMB 50098-50100. This was necessary as the inventory number ZMB 10100 had already been assigned to a specimen of "Mantidactylus granulatus" (= paralectotype of Limnodytes granulatus Boettger, 1881) from "Nosy Bé, don. Museum Senckenberg" (see Glaw and Vences 1993 (Barbour and Loveridge 1946: 128). Seventy-seven other paratypes, including specimens from Miramba, Ipiana, Transvaal and Lindi, as well as material collected by Wilms could not be located.  (Ahl, 1931a). Remarks. Photo in Ahl (1931b: 365, fig. 240, horizontally mirrored) showing the specimen on a leaf with 39 eggs. The type previously was regarded as lost, however, we rediscovered the specimen, still sitting on the leaf with the eggs (see Fig. 8). In accordance with Article 75.8 of the International Code of Zoological Nomenclature (ICZN 1999), the rediscovery of the holotype in the ZMB collection in 2012 renders the neotype designation by Perret invalid (1966: 410;MHNG 995.48 from "Foulassi, rivière Lobô"). Since 1988 the species was a member of the genus Alexteroon Perret, 1988. However, recently Ernst et al. (2021 revised the systematic position of Alexteroon and assigned the three species to the genus Hyperolius.

Hyperolius obstetricans
The German naturalist, botanist and gardener Zenker joined the German colonial service as taxidermist in 1889. He was manager of the colonial station Jaunde (Yaoundé, Mfoundi Department, Centre Region, Cameroon) from 1890-1895 (Zenker 1890). In 1896 he settled in Bipindi on the Lokundje River where he collected natural history and ethnological objects extensively and managed different plantations until his death on 6 Febru-ary 1922. The main part of his zoological collection is at ZMB (Mildbraed 1923;Frahm and Eggers 2001).  (Duméril & Bibron, 1841).
Remarks. Depicted in Ahl (1931b: 344, fig. 218). Another syntype, MCZ A-17643 with identical collecting data was sent in exchange to MCZ in 1932 (Barbour andLoveridge 1946: 130). Laurent (1961: 83) erroneously presumed that the five subadult specimens inventoried under ZMB 13701 could be the types of Hyperolius phrynoderma, but these specimens were collected at "Bukoba" by Stuhlmann and do not correspond with the locality given by Ahl for the types. Ahl's (1931a: 72) vague locality information for the H. phrynoderma types is probably wrong and mainly based on the transfer of the expedition name to a region, i.e. "Central German East Africa" which would be geographically equivalent to today's central Tanzania. However, the zoological collections of this expedition mainly originate from northeast Tanzania, Rwanda and the adjacent Democratic Republic of the Congo, more precisely from the region between Bukoba on the western shore of Lake Victoria, Ischangi in the south of Lake Kiwu and Irumu in the Ituru Province of northeastern Democratic Republic of the Congo (see comments in Barbour and Loveridge 1946: 130;map in Schubotz 1909).
Currently, the status of this taxon is uncertain. In comparison to species in the H. marmoratus / viridiflavus group, the body is comparatively slender and the snout more pointed. Dorsal warts are distinct, and arranged very regularly, even in the single adult frog (ZMB 39000). Juveniles of the H. marmoratus / viridiflavus group have warty skin, adults usually have smooth skin). Drewes (1997) described a superficially similar-looking frog from the Serengeti, Hyperolius orkarkarri, which is currently regarded as a synonym of H. glandicolor (see Channing and Howell 2006).

Hyperolius substriatus
Remarks. Ahl (1931a: 40) mentioned a series of 47 specimens from "Bukoba" including the "Type". We failed to locate 23 of these specimens. He also lists 11 additional specimens that we could not locate, as well as a second specimen from "Mohasi See", coll. Schubotz and two specimens without locality or collector information. Drawings of seven specimens showing the variation of this taxon are published by Ahl (1931b: 298, fig. 173). These have been copied from Tornier (1896, pl. 4, specimen no. 26-28 and 30-33). Another two paratypes, MCZ A-17648 and 17626, from "Bukoba" collected during the first "Deutsche Zentral- Afrika-Expedition", 1907 were sent in exchange to MCZ in 1932 (Barbour and Loveridge 1946: 129). Remarks. Hyperolius variegatus was described by Peters (1882a) from an unknown number of specimens originating from "Mocambique (Cabaḉeira, Quellimane, Inhambane)". Peters (1882b: 168) mentions that he first saw small specimens of this species on bushes on the "Cabaceira" peninsula in June 1843, an adult specimen on a mulberry tree in "Quellimane" in January 1846. He also lists an observation from "Prazo [estate] Boror" in March 1846, but did not mention the locality "Inhambane" in this second publication. Tornier (1896: 145, pl. 4, figs 61, 62) mentioned and depicted the two syntypes from "Quellimane, Mozambique", both inventoried together under ZMB 4530. These could not be traced by Bauer et al. (1995: 46), nor by us. The other syntype(s) from "Inhambane" with unknown inventory number could also not be located. Laurent (1961: 67) suggested that one of the specimens under 10249 is actually a specimen of Afrixalus fornasini (Bianconi, 1849). Our examinations revealed that ZMB 10249 und ZMB 75602 are H. marmoratus, whereas ZMB 75603 and ZMB 75604 are juvenile specimens of Afrixalus. Identification on the species level was not possible for the latter two frogs. Remarks. Tornier (1896, pl. 4, fig. 29) depicted the paralectotype; figure copied by Ahl (1931b: 338, fig. 211). Lectotype by subsequent designation through Laurent (1961: 88).

Megalixalus maculifer
However, the location "Quillimane" as mentioned in the ZMB inventory catalogue does not correspond to the type locality given by Pfeffer (1893). It is unclear whether this is a transmission error of the locality information for the ZMB specimens, and "Quillimane" instead of "Sansibar" is correct. At least for the Berlin syntype of Rappia vermiculata, the locality information has been mixed up, as shown by Tornier (1896: 141, see below).

Megalixalus uluguruensis
(ZFMK) and Jakob Hallermann (ZMH) for information on the availability of the publications published by Ahl 1931. Jakob Hallermann also provided information on the type specimens of the reed frogs described by J. G. Pfeffer and Silke Schweiger (NMW) sent us information on the amphibians and reptiles collected by R. Grauer kept in the NMW collection. Vasco Galante (Gorongosa National Park, Mozambique) provided some information concerning Mozambican localities. We are thankful to former and present staff of the ZMB library, Martina Rißberger, Hand-Ulrich Raake, Maria Xylaki, Vivien Bauer and the Department of Historical Research at ZMB, Sandra Miehlbradt, Sabine Hackethal, Hannelore Landsberg for procuring rare literature and access to archive material. We are grateful to Aaron Bauer, David C. Blackburn and Alan Channing for their thorough reading of a previous version of the manuscript. Their comments and constructive critique improved our paper!