L₁ larvae: 45 ex., 28.04.2012. Adults collected in Brelingen, north of Hannover (Germany), border of an oak forest, 80 m, 16.04.2012.

Mature larvae: 8 ex., 12.03.2016, Niedersachsen, Deister Mountains, Bredenbeck-Steinkrug, southwest of Hannover (Germany), 165 m (Fig. 40): 6 ex. between roots of young Fagus sylvatica L. and 2 ex. between roots of young Picea abies (L.) Karst. trees.

L₁ larvae: 80 ex., 10.06.2012. Adults collected in Brelingen from a row of old oak trees and from broom (Cytisus scoparius (L.) Link) in a disused sand pit nearby, 70 m, 15.04.2012.

Mature larvae (9 ex.): 1 ex., 04.10.2013, climate chamber of the Julius-Kühn-Institute in Braunschweig, breeding in flower-pots with Prunus laurocerasus L.; 4 ex., 17.04.2015, Sachsen-Anhalt, National Park Harz, Harz Mountains, Schierke, Hohnekamm, nutrient-poor grassland, 830 m; 1 ex., 30.03.2015, Sachsen-Anhalt, National Park Harz, Harz Mountains, Ilsenburg, Meineberg, pioneer forest with young birches (Betula pendula Roth), 340 m (Fig. 42); 2 ex., 06.07.2015, Niedersachsen, Brelingen, north of Hannover, disused sand pit, between roots of Cytisus scoparius, close to a row of old oak trees (Quercus robur L.), 70 m (Fig. 45); 1 ex., 30.12.2015, Niedersachsen, Hannover-Vahrenheide, Kugelfangtrift, nutrient-poor sandy grassland, 50 m, between roots of a mixed stand of Geranium pusillum L., Erodium cicutarium (L.) L’Hér. ex Aiton, grasses and a sedge species (Agrostis capillaris L., Carex hirta L., Festuca rubra L.), found together with pupae of Phyllobius pyri (Linnaeus) f. vespertinus (Fabricius).

L₁ larvae: 6 exx., 23.05.2012. Adults collected in Berkhof, north of Hannover, heathland and light pine forest with Calluna vulgaris, 10.05.2012.

Old larvae (2 exx.): 1 ex., 25.07.2012, collected from under roots of Calluna vulgaris in Berkhof (Fig. 44); 1 ex. and 1 immature adult, 02.09.2012, from breeding in a flower-pot with Calluna vulgaris in Hannover, Curculio Institute (Fig. 43). In these cases it is not known, whether the larvae were in the last or the penultimate instar.

Body (Figs 1, 3, 5). Moderately slender, curved, rounded in cross section. Prothorax slightly smaller than mesothorax; metathorax as wide as mesothorax. Abdominal segments I–VII of almost equal length. Abdominal segment VIII wide, flattened posteriorly, with conical lateral lobes. Abdominal segment IX strongly reduced, consisting of four, well isolated lobes, distinctly smaller than previous segments. Abdominal segment X consists of four anal lobes of almost equal size. Anus located ventrally (Figs 2, 4, 6). Chaetotaxy well developed, setae capilliform, variable in length. Each side of prothorax with 8–11 prns (pronotal setae) of unequal length; 2 ps (pleural s.) and 1 eus (eusternal s.). Meso- and metathorax (Figs 7, 18, 29) on each side with 1 prs (prodorsal s.), 4 pds, variable in length (postdorsal s.), 1 long as (alar s.), 2 ss (spiracular s.), variable in length, 1 long eps (epipleural s.), 1 ps and 1 eus. Each pedal area of thoracic segments with 6–7 pda (pedal s.), variable in length. Abd. I-VII (Figs 7, 8, 18, 19, 29, 30) on each side with 1 medium-length prs, 5 pds, various in length and arranged along the posterior margin of each segment, 1 short and 1 long ss, 2 eps and 2 ps, 1 lsts (laterosternal s.) and 2 short eus.Abd. VIII (Figs 9–11, 20–22, 31–33) on each side with 1 medium-sized prs, 3–4 pds, different in length and arranged along the posterior margin, 2 very short ss, paired eps and ps, 1 lsts and 2 short eus.Abd. IX (Figs 9–11, 20–22, 31–33) on each side with 2 ds (dorsal s.), medium in length, located close to the posterior margin of the segment, 1 medium ps and 2 short sts (sternal s.). Each vertical anal lobe (Abd. X) with a pair of minute setae, sometimes absent.

Head (Figs 12, 23, 34). Light yellow to dark brown, almost oval or suboval, frontal suture distinct, Y-shaped, endocarina absent. Setae on head capilliform. Des_{1, 2, 3, 5} (dorsal epicranial s.) usually equal in length; des₁ and des₂ located in the central part of epicranium, des₃ placed on frontal suture, des₅ located anterolaterally. Fs_{3, 4} (frontal s.) almost equal in length, fs₃ located anteromedially, fs₄ anterolaterally, close to epistoma. Les₁ and les₂ (lateral s.) equal in length, slightly shorter than des₁. Ves (ventral s.) short, poorly developed. Postepicranial area with 4 very short pes (postepicranial s.). A pair of small stemmata (st) located anterolaterally on each side of head. Antenna (Figs 13, 24, 35) located at the end of frontal suture; antennal segment with sensorium, reniform, located medially; basal membranous article with 2-6 basiconic sensillae. Labrum (Figs 14, 25, 36) almost semicircular, anterior margin rounded or slightly sinuously emarginate; 3 pairs of lrs (labral s.) of different length, lrs₁ placed medially, lrs₂ anteromedially, lrs₃ anterolaterally; all lrs without protuberances. Clypeus (Figs 14, 25, 36) of medium width, trapezium-shaped, lateral margins straight, anterior margin of clypeus straight or slightly concave; two pairs of cls (clypeal s.) reduced, vestigeal, located posteromedially; clss (clypeal sensorium) clearly visible, placed medially between cls. Epipharynx (Figs 14, 25, 36) with 3 pairs of rod-shaped or capilliform als (anterolateral s.) of almost equal length; 3 pairs of ams (anteromedial s.): ams₁ very short, ams₂ half the length of als, ams₃ shorter than ams₂; 2 pairs of finger-like mes (medial s.), variable in length: first pair placed anteriorly, second pair medially. Anterior margin of epipharynx smooth or serrate, due to the presence of thorn-like cuticular processes placed between labral rods. Labral rods rather elongated, converging posteriorly. Mandibles (Figs 15, 26, 37) slightly curved, narrow, with divided apex (teeth different in length). There is a protruding additional tooth on the cutting edge between the apex and the middle of the mandible; mds_1-2 capilliform, almost equal in length. Maxilla (Figs 17, 28, 39) with 1 stps (stipal s.) and 2 pfs (palpiferal s.) of equal length; mala with 8 finger-like or capilliform dms (dorsal malar s.), equal in size, and 4 vms (ventral malar s.); vms shorter than dms; mbs (malar basiventral s.) medium in length or short. Maxillary palpi with two palpomeres, basal with short mxps (maxillary palp s.); distal palpomere apically with a group of sensillae, each palpomere with a pore. Praelabium (Figs 16, 27, 38) almost rounded or heart-shaped with 1 long plbs (prelabial s.), located medially. Ligula with 1-2 capilliform ligs (ligular s.), variable in length. Premental sclerite clearly visible, Q-shaped. Labial palpi two-segmented; apex of distal palpomere with some sensillae; each palpomere with a pore. Postlabium (Figs 17, 28, 39) with 3 capilliform pslbs (postlabial s.), the first pair located anteromedially, the remaining two pairs laterally; pslbs₂ always distinctly longer than others.

The larva of Strophosoma (Neliocarus) faber distinguished according to Van Emden (1952) and Scherf (1964), the remaining species as in the present work. In S. sus based on the last or penultimate instar.

Subgenus Neliocarus

Subgenus Strophosoma

The data about the development of Strophosoma melanogrammum and S. capitatum (in part) have been mainly worked out and summarized by Grimm (1973) and Schauermann (1973). According to these sources both species develop in the Solling Mountains in the southern part of Niedersachsen (Germany) very similarly, and over a period of more than one year. After overwintering adults start to feed and to lay eggs from May to first half of July. Dieckmann (1980) found females with eggs still in the abdomen in August. The eggs are usually laid in clusters in any kind of sheltered situations, e.g. in cracks of cork, between dry leaves of litter, in soil (probably in crevices), between leaf or flower buds of the host plant, on the ground in needles of pine shoots (Breese 1948; Grimm 1973) and, in the laboratory, between rolled or pleated absorbent paper or between paper and the substrate. The eggs are held together by a viscid fluid. This agrees with observations in the laboratory, documented by Sprick and Stüben (2012), who found bulks of eggs deposited between the layers of absorbent paper close to its edges, where these were laid by the female’s ovipositor, and by the observations of Wolcott (1933) on neotropical Entiminae genera of the (same) former subfamily Brachyderinae. He had fixed double paper strips in the field on the top of stakes beneath cultivated young Citrus plants, and found many eggs deposited between the paper strips or sheets.

Larvae then develop in soil and feed mainly on fine roots until the third or last (fourth) instar and overwinter. In the following year the larvae complete their development and pupate, mainly in August. Adults of the new generation emerge from September to November and single specimens also in spring of the succeeding year. Adults climb for maturation feeding in tree crowns in spring, late summer and early autumn. They also feed in the herb layer and in leaf litter. Urban (1913) gave a strikingly different time of pupation for S. faber: the second half of May, producing a newly emerged weevil in June.

According to Schauermann (1973)S. melanogrammum and S. capitatum have 4 larval instars. But there is no information about how these instars were differentiated. The method of Sprick and Gosik (2014) for the determination of larval instars failed in these species, as the eyes are strongly protruding over the edge of the head in the pupae.

The breeding of Strophosoma melanogrammum and S. capitatum in the climate chamber, which was very successful in the case of Otiorhynchus larvae (Gosik et al. 2016), did not result in any S. capitatum larvae (two attempts) and only one larva of S. melanogrammum (four attempts), although host plants, the size of flower-pots and the soil substrate were varied. Also an addition of eggs bred before in keeping boxes was not successful. For these reasons no pupae could be obtained by breeding. Field-collecting was conducted only rarely in August, the main month of pupation.

The larval development of any other Strophosoma species is apparently unknown. Breeding of Strophosoma sus in two flower-pots with Calluna vulgaris and Erica arborea L. revealed the following data (but only in the pot with Calluna): egg-laying started in May (first adults being already active in April) and a newly emerged, weak, adult and a larva were obtained in the beginning of September; four further larvae were seen in the flower-pots, but a further check 3 or 4 weeks later did not reveal any larva, pupa or adult weevil. The reasons for the failure of the continuation of this breeding attempt are not known. Three pupae of Strophosoma cf. sus were found in the field at the heathland site near Berkhof with Calluna vulgaris and young Pinus sylvestris trees, where S. sus, S. capitatum and, rarely, also S. fulvicorne (Walton, 1846) were present (Gosik and Sprick 2013). Two pupae were found on 9^th July and one on 2^nd September, indicating an earlier start of pupation than in S. capitatum or S. melanogrammum from the Solling Mountains. This is not surprising, as the heathland near Berkhof with a population of S. sus is a warm and dry lowland area that allows earlier development. From this, it would be of interest to determine whether the development of S. capitatum or S. melanogrammum is also possible within one year at such warm sites, or whether larval overwintering and diapause are obligatory in these species.

Some basic information regarding the morphology of the larva of Strophosoma (Neliocarus) faber (Herbst, 1785), with drawings of head, mouthparts and apex of abdomen, can be found in the work of Urban (1913). However, the main sources of information about morphology (and egg-laying habits) of Strophosoma larvae are studies of Van Emden (1950, 1952), containing descriptions of the first larval instar of S. melanogrammum, S. capitatum and S. (Neliocarus) nebulosum Stephens, 1831 (syn. S. retusum Marsham, 1802), as well as of the mature S. faber larva. Van Emden (1952) also specified a number of characters of taxonomic importance for the entire genus Strophosoma, such as the presence of 3 or 4 pds on Abd. VIII, the conspicuously sclerotized apex of the abdomen, the shape of the premental sclerite, the proportions of setae on the pedal lobes, and the pds proportions on Abd. VIII.

The presence of four ventral malar setae in larvae of the genus Strophosoma confirms the observations of Marvaldi (1998a) that this character, as well as the shape of the antennal sensorium (Marvaldi 1997, 2003), can be considered an additional apomorphic character for larvae of the subfamily Entiminae, while other Curculionidae, e.g. Cyclominae in the sense of Marvaldi (1998b, 2003), possess the plesiomorphic number of five vms.

Further information on the morphology of preimaginal stages of the genus Strophosoma was provided by Willis (1964), who presented a very detailed, but unfortunately sparsely illustrated, description of the mature larva of S. melanogrammum. In addition to valuable information on the biology of reproduction and the morphology of larval stages of selected Entiminae species (genera Otiorhynchus Germar, 1822, Barynotus Germar, 1817, Sciaphilus Schönherr, 1823 and Strophosoma), Willis (1964) described a few general patterns valid for the morphology of weevils. Moreover, he observed changes in the proportions and relative lengths of setae when the larvae entered successive developmental stages. For this reason the key to the genus Strophosoma of Van Emden (1952), enabling identification of first-instar larvae to species level based on the proportions of setae, should be regarded – according to Willis (1964) – as at least inadequate for an identification of older larval instars. Changes in the proportions and shape of setae of weevil larvae when passing through successive developmental stages were also observed in later research on Entiminae (Gosik and Sprick 2012) and Hyperinae larvae (Skuhrovec 2004, 2006, 2007). Van Emden (1952) gave differences in the relative lengths of setae on pedal areas as the most important difference between Strophosoma sensu stricto and Neliocarus, but he did not mention the presence of four pds on the 8^th abdominal segment in Neliocarus versus three pds in Strophosoma.

The differences in the number of setae and in the morphology of abdominal segments which are observed in the species treated in this paper (S. capitatum, S. melanogrammum and S. sus) as well as in the previously described larva of S. faber (Van Emden 1952; Scherf 1964) show that Strophosoma species can be easily recognized in both mature stages and in younger larvae. The differences between them (e g. number of setae and presence or absence of asperities on the epipharynx) seem to support the existing division into subgenera (Strophosoma and Neliocarus). However, due to a lack of information about the morphology of the larvae of the remaining subgenera (Morphostrophus Flach, 1907, Pelletierius Alonso-Zarazaga & Lyal, 1999 and Subcaulostrophus Flach, 1907 according to Pelletier (2013)), a definitive evaluation of these characters will only be possible after further research.

Because of the morphology of the eighth abdominal segment, especially the sclerotization and the dark colour, larvae of the genus Strophosoma are among the most easily recognized weevil larvae. Only the larvae of the genera Philopedon Schönherr, 1826, and Tanymecus Germar, 1817, are similar to Strophosoma larvae with respect to the morphology of this segment (Van Emden 1950). This leads to the question: are these genera more closely related to each other than to other Entiminae genera? Here we can state the following: Smreczyński (1966) placed the genus Strophosoma in a separate tribe, Strophosomini, in a close relationship with the tribe Cneorhinini Lacordaire, 1863, with the genus Philopedon, whereas Tanymecus was placed in another subfamily (Tanymecinae). Dieckmann (1980) did not use tribes, but he also placed Philopedon next to Strophosoma (and between them only Cneorhinus Schönherr, 1823 with unknown larvae). Tanymecus was again placed in another subfamily. Today all these genera are members of the subfamily Entiminae, but they are placed in three different tribes: Brachyderini with Strophosoma (Pelletier 2013), Cneorhinini with Philopedon (Alonso-Zarazaga 2013) and Tanymecini with Tanymecus (Li Ren et al. 2013). A carefully reconstructed phylogeny, combining molecular, morphological and biological data, is still lacking.

Even if we do not know anything about the function of this conspicuous abdominal structure, the possibility that this is an adaptation to the environment has to be taken into account. It is found in a few genera of Entiminae only, larvae of which develop in soil. But there is no information as to whether these larvae exhibit behaviour or habits that are different from those larvae without a sclerotized 8^th abdominal segment.