(Ttt jf] «4S So V "01. Gr ■ y< ^ ANNALS MEDEDELINGEN 3 OF THE VAN HET TRANSVAAL MUSEUM VOLUME VI t 9 1334 ’ 'V'4> 30 2-OOj . /OlVAL Printed for the COMMITTEE OF THE TRANSVAAL MUSEUM BY THE UNIVERSITY PRESS, CAMBRIDGE, ENGLAND 1917-20 ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM VOLUME VI Printed for the COMMITTEE OF THE TRANSVAAL MUSEUM BY THE UNIVERSITY PRESS, CAMBRIDGE, ENGLAND 1917-20 PRINTED IN GREAT BRITAIN LIST OF CONTENTS VOLUME VI PAGE Hewitt, John Descriptions of New South African Araneae and Solifugae. (4 Plates and 13 Text-figures) 63 Meyrick, E., B.A., F.R.S. Descriptions of South African Micro-Lepidoptera ... 7 Pott, R. (Mrs) A New Species of Warburgia from the Transvaal. (2 Text-figures) . 60 Addendum to the First Check-list of the Flowering Plants and Ferns of the Transvaal and Swaziland. (2 Text-figures) . . . 119 Roberts, Austin Descriptions of a New Species and Genus of Flycatchers from East Africa, and two New Subspecies of Guinea Fowls from South Africa 1 Descriptions of two New Species of a New Subgenus of Bats, and a New Species of Molerats 4 Descriptions of some New Mammals 112 Some notes on Birds, and Descriptions of New Subspecies . . 116 INDEX TO VOLUME VI (New genera, subgenera, species, subspecies, and the main reference in a series of references in heavy-faced type; synonyms are in italics.) abrahami, Acanthodon, 76, hi Abutilon, 130 abyssinicum, Asplenium, 120 Acacia, 126 Acalypha, 129 ACANTHACEAE, 133 Acanthodon, 63, 67, 75, 102, 107, no, in acaulis, Geigeria, 135 Acompsia, 20 Acrocercops, 39 Acrostichum , 121 aculeastrum, Solanum, 133 aculeata, Aloe, 122 Eulophia, 124 Toddalia, 128 aculeatum, Polystichum, 120 acuminata, Markhamia, 133 acuta, Adenocline, 130 Elaeophorbia, 129 Lotononis calycina, 127 acutifolia, Melothria, 134 acutilobus, Cucumis africanus, 134 ADELIDAE, 47, 59 adendorffi, Mirafra apiata, 117 Adenia, 13 1 Adenocline, 130 Adiantopsis, 121 aegyptiacus, Nyctinomus, 4, 5 aethiopica, Zantedeschia, 122 affinis, Aloe, 122 Cluytia, 129 Tragia, 130 africana, Dryopteris, 120 Plukenetia, 130 Salvia, 133 africanum, Psilotrichum, 125 Thamnosma, 128 AIZOACEAE, 12s alata, Pentatricha, 134 Albizzia, 126 albogularis, Crithagra, 116 albomaculata, Zantedeschia, 122 Alchemilla, 126 Alepidea, 132 allevata, Hyalochna, 30 Allotalanta, 30 Aloe, 122 AMARANTACEAE, 125 AM ARYLL I D ACE AE, 122 Ammannia, 13 1 Amphilophis, 121 amphizeucta, Lachnostola, 22 amplexicaule, Ecbolium, 133 Anacampsis, 19, 20 ANACARDIACEAE, 130 Anaglypha, 134 Anarsia, 21 Ancylometis, 27 Ancylotrypa, 108, 109 Andropogon, 121 angulatum, Abutilon, 130 angustata, Acalypha, 129 angustifolia, Merremia, 132 Anisotes, 133 anomalus, Georychus, 5 ANONACEAE, 126 antennifera, Bonatea, 123 anthelmintica, Albizzia, 126 Anthericum, 122 Antispila, 35 aphrocyma, Acrocercops, 39 apiata, Mirafra, 117 aponeurus, Hibiscus, 130 Aponogeton, 121 APONOGETONACEAE, 121 Apotactis, 52 appro ximans, Euplectes, 117 aquilina , Pteris, 121 aquilinum, Pteridium, 121 arabica, Psiadia, 134 arabicus, Lotus, 128 Arabis, 126 ARACEAE, 122 ARANEAE, 63, 67 areata, Laspeyresia, 13 arenophilum, Tryphostemma, 13 1 arguta, Blastobasis, 36 Argyrolobium, 127, 128 Argyroploce, n, 12, 50 argyrozona, Antispila, 35 aristata, Chaetacme, 124 Lotononis , 127 Pearsonia, 127 aristatum, Aspidium, 120 Polystichum, 120 Aristotelia, 15, 16 armatus, Spiroctenus, 86, 87 (Text-fig. 7 d) Arthrosolen, 13 1 artifex, Stasimopus, 95 ASCLEPIADACEAE, 132 asiatica, Chrysochloris, 113, 114 aspera, Eragrostis, 122 Geigeria, 135 asperata, Tinea, 44 aspidioides, Asplenium, 120 2 Annals of the Transvaal Museum Aspidium, 120 Asplenium, 120 assimilis, Pavetta, 134 asterias, Thesium, 125 asterota, Argyroploce, 12 Astragalus, 128 athamantica, Dryopteris, 120 athamanticum , Nephr odium, 120 athanasioides, Brachymeris, 135 Pentzia , 135 Atherstonei, Pearsonia, 127 Satyrium, 123 Athyrium, 120 atrisecta, Hapsifera, 58 atropurpurea, Pupalia, 125 attenuata, Cephalaria, 134 Lomaria , 120 attenuatum, Blechnum, 120 augur, Rhinolophus, 112 Aulotropha, 32 aurantiaca, Sandersonia, 122 aurantiacus, Gladiolus, 123 aurea, Gymnogramme, 120 Gymnogramme argentea, 120 auriculata, Tripteris, 135 aurita, Blumea, 134 austerodes, Chelaria, 22 australis, Crotalaria petiolaris, 127 Evagrus caffer, 63, 95 (Text-figs. 10 a, b) Thelechoris, 95, 109 autograpta, Leucoptera, 40 automorpha, Syrmadaula, 26 Autosticha, 26 AVICULARIIDAE, 63, 67 azorella, Muraltia, 129 babaeculus, Bradypterus, 116 Bainesii, Lotononis, 127 Bakeri, Eulophia, 124 Barberton, Crassula perfoliata, 126 Barbertoni, Habenaria, 123 Barbertoniae, Aloe, 122 barratti, Bradypterus, 116 BARY CHELIDAE, 97 BARY CHELINAE, 96 basinuda, Alepidea, 132 Batrachedra, 28 Baurii, Ranunculus, 126 Bellidiastrum, Venedium, 135 henedicta, Ceromitia, 47 Tacazzea apiculata, 132 Benthami, Acacia, 126 Bergia, 13 1 bergiana, Dryopteris, 120 bergianum, Nephrodium , 120 Berkheya, 135 Berkheyopsis, 135 Bersama, 130 Bessia, 86, 88, 107 biaurita, Pteris, 121 bicolor, Habenaria, 123 bifurcum, Solanum, 133 BIGNONIACEAE, 133 biharicus, Idiops, no biloba, Marsilia macrocarpa, 121 binotata, Lecithocera, 24 bipinnatum, Asplenium, 120 bisecta, Eucosma, 10 bivia, Dragmatucha, 25 bivittatus, Diores, 105, 106 (Text-fig. 13 d) BLASTOBASIDAE, 36, 55 Blastobasis, 36, 55 Blechnum, 120 Blepharis, 133 Blossia, 63, 64 Blumea, 134 Boltoni, Bonatea, 123 Bolusii, Euphorbia, 129 Pleiospora, 127 Bonatea, 123 BORAGINEAE, 132 boryana, Lomaria, 120 Bovinae, in Bowkeri, Pelargonium, 128 Brachiaria, 122 Brachionopus, 96 Brachmia, 25, 26 brachycephalus, Ceratogyrus, 63, 103; PI. I, figs, a-c Brachycorythis, 123 br achy cor y this, Platanthera, 123 Brachymeris, 135 brachyphyllus, Gladiolus, 123 brachypterus, Bradypterus, 116 bracteata, Tricalysia, 133 Bradypterus, 116 Brassica, 126 breijeri, Pelmatorycter, 92 Brenthia, 36 brevepedunculata, Pleiospora obovata, 127 brevifolius, Gladiolus varius, 123 breyeri, Pelmatorycter, 63, 91 (Text- fig. 9); PI. II, fig. c Pterinochilus, 63, 102 Breyeri, Warburgia, 60 (Text-fig. 1), 61 (Text-figs. 2 a-g), 13 1 Bridelia, 129 brizantha, Brachiaria, 122 brizanthum, Panicum , 122 Brownleea, 123 Bucculatrix, 42 Buchanani, Dryopteris, 120 Nephrodium, 120 Buchenroedera, 128 Bulbostylus, 122 bullifera, Platyptilia, 7 Bupleurum, 132 burkeanus, Astragalus, 128 Burkei, Thesium, 125 Index to Volume VI 3 BURSERACEAE, 128 byrsoxantha, Trichotaphe, 23 Ceromitia, 47, 59 Ceropegia, 132 Cadaba, 126 caerulea, Sutera, 133 caffer, Evagrus, 95, 109 Heligmomerus, 63, 78 caffra, Ximenia, 125 caff rum, Erythroxylon, 128 cajanifolia, Pleiospora, 127 Cajanus, 11 calanthoides, Eulophia, 124 calomelanos, Ceropteris, 121 Pellaea , 121 Calostephane, 135 Camara, Lantana, 133 CAMPANULACEAE, 134 CANELLACEAE, 13 1 canescens, Cineraria, 135 canoargentea, Lasiosiphon, 13 1 capense, Blechnum, 120 Xanthoxylon , 128 capensis, Adiantopsis, 121 Cheilanthes, 121 Cissus, 130 Ctenomeria, 130 Eptesicus, 1 13 Euplectes, 117 Fagara, 128 Fleurya, 124 Hemitelia, 120 Plumbago, 41 Trachypogon polymorphus, 121 caperonioides, Acalypha, 129 capitata, Hembstaedtia, 125 Jacquemontia, 132 capnias, Parectopa, 40 CAPPARIDACEAE, 126 Capparis, 126 cardinata, Cholotis, 28 cardiophora, Disperis, 124 Carduus, 135 carnosa, Huernia, 132 Carposina, 8 CARPOSINIDAE, 8 Carteri, Eretmocera, 35 caryocoma, Argyroploce, 11 CARYOPHYLLACEAE, 125 Cassia, 127 castaneus, Stasimopus, 94 cathartica, Bridelia, 129 catopteron, Nephrodium, 120 caulescens, Dimorphotheca, 135 caulota, Labdia, 27 CAVICORNIA, in cepapi, Paraxerus, 112 Cephalaria, 134 Ceranthes, 33 Cerastium, 125 Ceratogyrus, 63, 103 ceresiiforme, Monocymbium, 121 Ceropteris, 121 certa, Oinophila, 57 certificate, Sapheneutis, 45 Ceterach, 120 Chaetachme, 124 Cheilanthes , 121 Chelaria, 21, 22 chilocalyx, Cleome, 126 Chirocompa, 30 Chloropeta, 1 Chloropetella, 1 chlorotoma, Leuronoma, 16 Onebala, 25 Cholotis, 28 chrysocarpus, Rubus rigidis, 126 Chrysochloris, 113 cicutarium, Aspidium, 120 ciliata, Acalypha, 129 ciliolata, Peperomia retusa, 124 Cineraria, 135 cineria, Vernonia, 134 cir cum junc turn, Galeosomavandami, 63, 78 (Text-fig. 4c), 80; PL IV, figs. /, g cirrhocoma, Anacampsis, 20 Cissus, 130 Cistugo, 1 12 citri, Prays, 56 Citrus, 56 civetta, Viverra, 112 Cladophantis, 33 Clarkei, Blepharis, 133 clavigera, Euphorbia, 129 Cleome, 126 Cloeotis, 1 12 Cluytia, 129 Cnephasia, 9 Cnetis, 128 Coccolus, 126 coegensis, Moggridgea, 107 COLEOPHORIDAE, 38 coloratus, Sporobolus pectinatus, 122 combreticola, Viscum, 124 Commiphora, 128 comosa, Alepidea, 132 Hermannia, 13 1 compacta, Depressaria, 31 COMPOSITAE, 134 ccmcinna, Davallia , 120 conclusa, Trichotaphe, 23 concolor, Doryopteris, 121 confixa, Telphusa, 51 conjuncta, Muraltia, 129 CONNARACEAE, 128 Conrathii, Senecio, 135 conservata, Protomacha, 31 consobrina, Pellaea, 121 constrictos, Pterinochilus, 102 ftn. convallata, Myrmecozela, 57 4 Annals of the Transvaal Museum CONVOLVULACEAE, 132 Conyza, 134 Cooperi, Crassula, 126 Ranunculus, 126 COPROMORPHIDAE, 35 Corallocarpus, 134 Corchorus, 130 cordata , Gymnogramme, 120 cordatum, Ceterach, 120 cordifolium, Schizoglossum, 132 Coreopsis, 135 Coriacea, Rhus, 130 coronata, Numida mitrata, 3 coronatum, Galeosoma, 78 (Text-fig. 4c), 81, 82, 83; PL IV, fig. a corroborata, Tortrix, 8 corrugata, Gracilaria, 40 COSMOPTERYGIDAE, 27, 52 costatum, Thesium, 125 Cotyledon, 126 cousinioides, Berkheya, 135 Crabbea, 133 crassispina, Pterinochilus, 100, 102 Crassula, 126 CRASSULACEAE, 119, 126 crataegifolia, Vernonia,, 134 crateraula, Pycnostola, 14 cregoei, Acanthodon, 75 crenata, Dryopteris, 120 crenatum , Nephrodium , 120 crinitum, Galeosoma robertsi, 63, 82; PI. IV, fig. b crispula, Harveya, 133 Crithagra, 116 Crobylophora, 41 Crocidura, 112 Crossandra, 133 Crotalaria, 127 Croton, 129 CRUCIFERAE, 126 crudeni, Acanthodon, 72, 76, 77 Pelmatorycter, 108 Crypsithyris, 43, 57 crypsixantha, Nepticula, 43 Cryptillas, 116 CTENIZIDAE, 107 ftn., m Ctenolophus, 75, no Ctenomeria, 130 cuculans, Odites, 54 Cucumis, 134 CUCURBITACEAE, 134 culveri , Disa, 124 curvipes, Spiroctenus, 63, 88, 89 (Text- figs. 8 a-c); PI. Ill, figs, d-e cyanea, Crocidura, 112 cyanoscia, Zesticodes, 46 CYATHEACEAE, 120 cymatias, Microschismus, 35 Cymbopogon , 121 CYPERACEAE, 122 Cyphia, 134 Cyphothyris, 32 cyprophanes, Daemonarcha, 27 Cyrtomium, 120 Cystopteris, 120 cytisoides, Thesium, 125 Daemonarcha, 27 Dalbergia, 128 Dalechampia, 130 Damarchodes, 109 damarensis, Numida papillosa, 2, 3 daricella, Crobylophora, 41 darlingi, Ceratogyrus, 103 Rhinolophus, 112 Davalia, 132 Davallia, 120 Davyae, Musa, 123 Davyana, Aloe, 122 Davyi, Fagara, 128 decachrysa, Glyphipteryx, 36 Decadarchis, 43 decurrens, Chirocompa, 30 deltophanes, Porthmologa, 53 densiflora, Dombeya, 13 1 dentata, Pteris, 121 dentatus, Senecio Johannesburgensis, 135 denticulata, Streptocarpus, 133 depauperata, Nidorella, 134 depressa, Selaginella, 121 Depressaria, 31 depressinervia, Acalypha, 129 derogatella, Eretmocera, 35 deserti, Sanseviera, 122 desipiens, Eucosma, 10 Desmodium, 128 diagonalis, Digitaria, 12 1 Dianthus, 125 dichiloides, Lotononis, 127 Dichomeris, 23 Dichondra, 132 dichotoma, Gleichenia, 120 dichotomum, Viscum, 124 Dichrocephala, 134 Dicliptera, 133 Digitaria, 121 Dimorphotheca, 135 Dinteri, Crotalaria squarrosa, 127 Ficus, 124 Diocosma, 33 dioecus, Scirpus, 122 Diores, 63, 105 Dioscorea, 123 DIOSCOREACEAE, 123 diplopsamma, Sapheneutis, 45 DIPLOTHELEAE, 107 ftn. DIPLURIDAE, 107 ftn. DIPSACEAE, 134 Disa, 124 discolor, Stephania, 126 discoporus, Sporobolus, 122 Disperis, 124 Index to Volume VI 5 Disperma, 133 dissectus, Cucumis hirsutus, 134 Dissotis, 13 1 distans, Crotalaria, 127 divaricata, Calostephane, 135 Raphionacme, 132 diversifolius, Hibiscus, 130 dolichocephalus, Ceratogyrus, 63, 104; PI. I, fig. d\ PL II, fig. a Dombeya, 13 1 dorsistrigata, Eretmocera, 35 Doryopteris, 121 Dragmatucha, 24, 25 drastica, Eucosma, 49 dregeana, Dioscorea, 123 Thunbergia, 133 Vernonia, 134 dregeanum, Cerastium, 125 Desmodium, 128 dreyeri, Stasimopus, 93 drimylota, Apotactis, 52 dryadopa, Brachmia, 25 Dryodromas, 117 Dryopteris, 120 dubia, Leptopelma, 97 dura, Pellaea, 121 Dyschoriste, 133 Ecbolium, 133 Eclipta, 135 edouardi, Guttera, 3 effulgens, Limnoecia, 27 effusum , Sorghum halepense , 121 egens, Blastobasis, 37 Elachista, 55, 56 ELACHISTIDAE, 55 elaeocoma, Anacampsis, 19 Elaeophorbia, 129 Elaphoglossum, 121 ELATINACEAE, 13 1 elatus, Gladiolus varius, 123 elongata, Dryopteris, 120 elongatum , Nephrodium Filix-Mas , 120 elongatus, Felicia, 134 embolaea, Hemimene, 51 empetroides, Muraltia, 129 encharacta, Argyroploce, 50 enormis, Euphorbia, 129 eodryas, Acompsia, 20 eotrocha, Diocosma, 33 epapposum, Helichrysum, 134 EPERMENIADAE, 56 Ephippias, n epicoena, Proterochyta, 56 epicyparissias, Euphorbia, 129 Epimys, 112 Epiphractis, 53 Epithectis, 16 Eptesicus, 1 13 Eragrostis, 122 erebaula, Pycnodytis, 15 erecta, Eclipta, 135 erectum, Asplenium, 120 Eretmocera, 35 Eriochrysis, 121 eriophorum, Helichrysum, 134 Erlangea, 134 ermelensis, Disperis, 124 Erotis, 33 erythropa, Argyroploce, 11 Erythrorrhizum, Anthericum, 122 ERYTHROXYLACEAE, 128 Erythroxylon, 128 Ethelae, Kaempferia, 123 Ethmia, 37 Eucosma, 9, 48 EUCOSMIDAE, 9, 48, 51 Eucryptogona, 45 Eugenia, 13 1 Eulalia, 121 Eulophia, 124 Euphorbia, 129 EUPHORBIACEAE, 119, 129 Euplectes, 117 euplocamis, Tinea, 44 euryacta, Lysitona, 57 euryzancla, Oxymachaeris, 43 Euxanthis, 8 Evagrus, 63, 95, 109 Evansii, Euphorbia, 129 excoriata, Laspeyresia, 13 exhilarata, Argyroploce, 50 exoenota, Gelechia, 52 explicata, Digitaria monodactyla, 121 exsanguis, Carposina, 8 exsulata, Anacampsis, 20 extensa, Blastobasis, 55 Fagara, 128 falcatum, Aspidium, 120 Cyrtomium, 120 Farm Goede Hoop, Geranium incanum, 128 fecunda, Gelechia, 17 Felicia, 134 ferox, Acacia, 127 ferulata, Parapsectris, 17 ficifolia, Pueraria, 128 Rhynchosia, 128 Ficus, 124 filamentosa, Crassula, 126 filifolia, Lotononis, 127 Pearsonia, 127 filiformisy Rhus, 130 fiscinata, Brachmia, 26 flabellatay Pteris, 121 FLACOURTIACEAE, 13 1 flaveolum, Acanthodon, 77 and ftn., 107 flavisecta, Cnephasia, 9 flavivittus, Heliosciurus, 112 flavopunctatus, Hermachastes, 107 Fleurya, 124 6 Annals of the Transvaal Museum floccosum, Helichrysum, 134 floribunda, Tricalysia, 133 floribundum, Thesium, 125 Fluggea, 129 foliosa, Habenaria, 123 Forskalii, Themeda, 121 fossoria, Bessia, 86 fragilis, Cissus, 130 Cystopteris, 120 fulvicapilla, Dryodromas, 117 Fumaria, 126 fumarioides, Hymenophyllum, 120 Fumea, 45, 59 furcatum, Asplenium, 120 fuscipennis, Eretmocera, 35 Galeosoma, 63, 78 Galpini, Acalypha caperonioides, 129 Euphorbia, 129 Habenaria, 123 Galpinii, Ficus, 124 Loranthus, 124 garcianus, Loranthus, 124 Gardenia, 133 Geigeria, 135 Gelechia, 17, 18, 52 GELECHIADAE, 13, 51 geoffroyi, Rhinolophus, 112 geomicta, Phthorimaea, 18 Georychus, 5 GERANIACEAE, 128 geraniaefolis , Pellaea , 121 Geranium, 128 Gerbera, 135 Gerdana, 32 Gerrardi, Schizochilus, 123 GESNERACEAE, 133 Gettleffii, Stapelia, 132 Giesekia, 125 glabra, Amphilophis, 121 Vernonia, 134 glabrata, Acalypha, 130 Leucas, 133 glabrescens, Nesaea sagittifolia, 13 1 glabricarpellata, Knowltonia, 125 glabriflorus , Loranthus, 124 Gladiolus, 123 glandifera, Ethmia, 37 glandulifera, Wormskioldia, 13 1 glauca, Adenia, 13 1 glaucella, Euphorbia, 129 Gleichenia, 120 GLEICHENIACEAE, 120 globuligemma, Aloe, 122 glomeratum, Limeum, 125 glyphicodes, Eucosma, 10 GLYPHIPTERYGIDAE, 36, 55 Glyphipteryx, 36 Gnidia, 13 1 godfreyi, Diores, 63, 105, 106 (Text- figs. 13 a-c) goetzeanum, Thesium, 125 gongylodes, Dryopteris, 120 Gorgyrella , 75, no, in Goudotii, Pellaea, 121 Gracilaria, 40 GRACILARIADAE, 39 gracilarioides, Thesium, 125 gracile, Asplenium lunulatum, 120 gracilior Pleiospora, 127 gracilipes, Acanthodon, 63, 68, 75 gracilirostris, Lusciniola, 116 gracilis, Juncus, 122 gracillima, Rhus, 130 Wahlenbergia, 134 gracillimus, Podocarpus, 121 grahami, Epimys namaquensis, 112 GRAMINEAE, 121 grandiflora, Tavaresia Barklyi, 132 grandis, Mungos, 112 grantiana, Crotalaria, 127 gratiosa, Simaethis, 55 gratissimus , Croton, 129 Grewia, 130 griquensium, Scirpus, 122 grisea, Hermannia, 13 1 grosseserrata, Hermannia, 13 1 Gubouga, Croton, 129 Gueinzii, Euphorbia, 129 Guilleminea, 125 guineense, Erythrophloem, 127 guineensis, Sanseviera, 122 gurneyi, Proctopus nigricollis, 118 Guttera, 3 Gymnogramma, 38 Gymnogramme, 120, 121 gypsophiloides, Thesium, 125 haagneri, Platymops, 5 Habenaria, 123 HALORHAGIDACEAE, 132 hamatum, Mesembrianthemum, 125 Hapsifera, 46, 58 Harpactirella, 96, 97 harpalea, Opogona, 56 Harveya, 133 harveyana, Vitex, 133 hastata, Pellaea, 121 Heeria, 130 Helichrysum, 134 Heligmomerus, 63, 78 HELIODINIDAE, 35 Heliosciurus, 112 Heliotropium, 132 HELIOZELIDAE, 35 helminthias, Schiffermuelleria, 29 Hemimene, 51 Hemitelia, 120 hemizona, Leucoptera, 41 hepburni, Acanthodon, 63, 73, 77 Hermacha, 109 Hermachastes, 107 Index to Volume VI 7 Hermannia, 13 1 Hermbstaedtia, 125 Hesperarcha, 38 heteracantha, Acacia, 127 Hexalobus, 126 Hibiscus, 130 Hippocratea, 130 HIPPOCRATEACEAE, 130 hirsuta, Cluytia, 129 hirsutissima, Lotononis calycina, 127 hirsutum, Galeosoma, 79, 82 Thesium, 125 hirsutus, Acanthodon, 63, 69, 75 Corchorus, 130 Gnidia fastigata, 13 1 hirta, Euphorbia, 129 Hyparrhenia, 121 Mollugo, 125 Nidorella, 134 hirtus, Cymbopogon, 121 hispida, Polygala, 129 hoepfnerianus, Lasiosiphon, 13 1 holosericea, Cassia, 127 Pleiospora, 127 holosticta, Ceromitia, 59 Holothrix, 123 homogramma, Onebala, 25 Homostola, 90, 108 horizontalis, Digitaria, 121 horrida, Acacia, 126 hortulana, Trichotaphe, 23 hostilis, Polyhymno, 19 Huernia, 132 humilior, Lotononis, 127 Hyalochna, 30 HYDROCHARITACEAE, 121 Hygrophila, 133 HYMENOPHYLLACEAE, 120 Hymenophyllum, 120 Hyparrhenia, 121 hypericifolia, Euphorbia, 129 Hypolepis, 121 hypoleuca, Sanguisorba, 126 HYPONOMEUTIDAE, 37, 56 hysterota, Platybathra, 38 idiocoma, Opostega, 42 Idioglossa, 56 IDIOPEAE, hi Idiops, no, hi Idiothele, 63 , 96 illecebrioides, Guilleminea, 125 illuminata, Pycnostola, 15 imbertis, Sisyranthus, 132 inaemoena, Eulophia, 124 inaequalis, Crassula, 126 Dryopteris, 120 inaequilatera, Euphorbia, 129 incana, Rhus, 130 incanum, Geranium, 128 Polypodium , 121 incisifolia, Tragia, 130 inclusus, Gladiolus, 123 incolumis, Odites? 54 incurva, Habenaria, 123 Indigofera, 128 ingens, Euphorbia, 129 Ficus, 124 inscita, Eucosma, 9 insculpta, Amphilophis, 121 insolita, Crypsithyris, 57 intorta, Polyhymno, 19 involucrata , Euphorbia, 129 iocharis, Trachydora, 28 Ipomoea, 132 IRIDACEAE, 123 irroratus, Otomys, 115 Isachne, Brachiaria, 122 Panicum , 122 isatideus, Senecio, 135 Ischnothele, 96 Jacobsziae, Alepidea, 132 Jacquemontia, 132 Jasminum, 132 Jatropha, 129 Jenkinsii, Alepidea, 132 JUNCACEAE, 122 Juncus, 122 junodi, Pterinochilus, 102 ftn. Junodiana, Brachycorythis, 123 Junodii, Kalanchoe, 126 Thesium, 125 Kaempferia, 123 kalachariensis, Loranthus, 124 Kalanchoe, 126 Karroo, Acacia, 126 karschi, Ischnothele, 96 kentanicus, Acanthodon, 68, 69 Kirkii, Dalechampia, 130 Loranthus, 124 Kniphofia, 122 Knowltonia, 125 komatiensis, Rhynchosia, 128 kraussiana , Acacia arabica, 126 Kraussiana, Euphorbia, 129 kraussianus , Loranthus, 124 Labdia, 27, 52 labialis, Tetramnus, 128 LAB I AT AE, 133 Lachnostola, 22 lacunosa, Epithectis, 16 laeta, Disa, 124 Lagerosiphon, 121 laminatus, Otomys, 114, 115 laminicornis, Blossia, 63, 65, 66 (Text- figs. 2 a-c) lamprostola, Monopis, 43 lanceolata, Crotalaria, 127 Gymnogramme , 121 8 Annals of the Transvaal Museum lanceolata, Lotononis, 127 Lantana, 133 lanuginosa, Dryopteris, 120 Lasiosiphon, 13 1 Lasiospermum, 135 Laspeyresia, 12, 51 latebracteolata, Pleiospora, 127 laticosta, Blossia, 63, 64 (Text-figs. 1 a-c) latifolia, Acalypha glabrata, 130 Anaglypha, 134 Ceropegia multiflora, 132 Dichrocephala, 134 latifolium, Elaphoglossum conforme, 121 latipetala, Eulophia, 124 latissimifolius, Senecio, 135 laxa, Cluytia, 129 Lecithocera, 34 ledereriella, Batrachedra, 28 Leendertziae, Loranthus olaefolius, 124 LEGUMINOSAE, 11, 119, 126 Leptopelma, 97 Lessertia, 128 lesueuri, Cistugo, 112 Leucas, 133 leucatoma, Brenthia, 36 Leucoptera, 40 leucoptera, Crithagra, 116 Leuronoma, 16 LILIACEAE, 122 Limeum, 125 Limnoecia, 27 linearis, Gleichenia, 120 linoides, Gnidia, 13 1 Lissochilus, 124 lividicollis, Guttera, 3 lobatum, Asplenium erectum , 120 lobostola, Eucosma, 49 Lomaria , 120 lomatophyllus, Juncus, 122 londinensis, Spiroctenus, 63, 86, 87 (Text-figs. 7 a-b), 107 longeciliata, Alepidea, 132 longibracteata, Aloe, 122 longiflora, Digitaria, 121 Thorncroftia, 133 longifolia, Acalypha punctata, 130 longifolium, Argyrolobium, 127 Lopholaena, 135 LORANTHACEAE, 119, 124 Loranthus, 124 Lotononis, 119, 127 Lotus, 128 loxogramme, Polypodium, 121 loxosaris, Chelaria, 21 luctuosum, Polystichum, 120 lugardae, Commiphora, 128 Monadenium, 129 Lugardii, Sesamothamnus, 133 lunariifolius, Hibiscus, 130 lunifera, Eretmocera, 35 lunulatum, Asplenium, 120 Lusciniola, 116 lutea, Felicia, 134 luticoma, Fumea, 59 lydenbergensis, Monodenia, 124 LYONETIADAE, 40, 42, 56 lyrata, Cineraria, 135 Lysitona, 57 LYTHRACEAE, 13 1 MacLeaii, Aspidium, 120 Polystichum, 120 macowanii, Hermannia, 13 1 macrobela, Labdia, 52 macrochilus, Pachycarpus, 132 macrophylla, Oldenlandia, 133 Pleiospora, 127 macrophyllus, Senecio, 135 macropoda, Tephrosia, 128 macrorhynchus, Euplectes capensis, 117 Maerua, 126 magaliesmontanus, Gladiolus, 123 magalismontana , Rhus, 130 magalismontanum, Thesium, 125 magna, Harpactirella, 97 magnusiana, Ipomoea, 132 major, Lagerosiphon muscoides, 121 mallotocarpa, Ficus, 124 MALVACEAE, 130 marginata, Lotononis , 127 Pearsonia, 127 marginatus, Gladiolus, 123 maritimus, Mungos pulverulentus, 114 Markhamia, 133 Marleyi, Spiroctenus, 63, 83 (Text- fig- 5) Marlothii, Hymenophyllum, 120 Lotononis, 127 marshalli, Ceratogyrus, 104 Marshallii, Aloe, 122 Marsilia, 121 MARSILIACEAE, 121 Matricaria, 135 mauritiana, Dryopteris, 120 maxillaris, Crotalaria, 127 melanaula, Eucosma, 11 melanogastra, Cholotis, 28 melanoleuca, Zantedeschia, 122 melanostola, Phrixosceles, 39 melanoxylon, Dalbergia, 128 Melasina, 45 MELASTOMACEAE, 13 1 melckorum, Eptesicus, 113 Melhania, 13 1 MELIACEAE, 128 MELIANTHACEAE, 130 melitardis, Opostega, 41 Melolobium, 127 Melothria, 134 Index to Volume VI 9 MENISPERMACEAE, 126 merista, Pycnostola, 14 Merremia, 132 Mesembrianthemum, 125 mesochlora, Talaeporia, 44 mespilifolia, Vernonia, 134 METACHANDIDAE, 27 metapyrrha, Tortrix, 8 microcarpa, Chaetachme, 124 Fluggea, 129 Microcolona, 53 Microlepia, 120 MICROLEPIDOPTERA, 7, 48 micropetalus, Dianthus, 125 microps, Acanthodon, 69, 75, 111 Microschismus, 35 MIGIDAE, 107 ftn. miniata, Eretmocera, 35 minor, Acanthodon schreineri, 76, 1 1 1 Bessia, 86, 88 Chrysochloris, 113 Spiroctenus, 87 (Text-fig. 7 c) Stasimopus, 93 minus, Thalictrum, 126 minuscula, Decadarchis, 43 Mirafra, 117 mitrata, Numida mitrata, 3 Moggridgea, 107 molle , Nephr odium, 120 mollis, Bridelia, 129 Dryopteris, 120 Mollugo, 125 Monactinocephalus, 134 Monadenium, 129 Monechma, 133 Monocymbium, 121 Monodenia, 124 Monopis, 43 montana, Brachymeris, 135 montanum, Trichomanes, 120 Monteiroi, Crotalaria, 127 monticola, Acanthodon, 67, 76 Cluytia, 129 monticoloides, Acanthodon, 63, 67, 76 Moorei, Loranthus, 124 MORACEAE, 124 mossambicum, Galeosoma, 63, 78 (Text-figs. 4 a-b) mossambicus, Acanthodon, 63, 72, 73, 76 Mucuna, 128 Muddii, Argyrolobium, 127 multiflora, Adenia, 13 1 Lotononis, 127 Lotononis laxa, 127 Matricaria, 135 Pearsonia, 127 Wahlenbergia, 134 Mundtii, Bupleurum, 132 Fumaria, 126 Scolopia, 13 1 Mungos, 1 12, 1 14 munroanum, Saccharum , 121 Muraltia, 129 muricata, Sanguisorba, 126 Musa, 123 MUSACEAE, 123 Myotis, 1 12 myriacantha, Aloe, 122 Myrmecozela, 57 MYRTACEAE, 13 1 namaquensis, Blossia, 65 Loranthus, 124 natalense, Aponogeton, 121 Thesium, 125 natalensis, Cadaba, 126 Chloropeta, 2 Cnetis, 128 Euphorbia, 129 Georychus, 6 Gerbera, 135 Spermacoce, 134 Toddalia, 128 Tragia, 130 Ursinia, 135 natalitia, Crotalaria, 127 natalitium, Desmodium, 128 natans (obovata), Crassula, 126 Nationae, Thesium, 125 Nebolusia, 123 Nelsoni, Brownleea, 123 Disa, 124 Eulophia, 124 Nelsonii, Disperis, 124 neopolycnemoides, Euphorbia, 129 neoxesta, Depressaria, 31 Nephr odium, 120 Nepticula, 43 NEPTICULIDAE, 43 neritis, Phyciodyta, 58 Nesaea, 13 1 Nidorella, 134 nigellaefolia, Matricaria, 135 nigellus, Stasimopus, 63, 93 nigra, Hermacha, 109 Moggridgea, 107 nigrispersa, Scythris, 37 nigrofulvus, Idiothele, 63, 98, 100 (Text- fig. 11), 101 (Text-fig. 12 a), 102; PI. II, fig- b Pterinochilus, 98 nigropedata, Brachiaria, 122 nigropedatum, Panicum, 122 nigropilosus, Acanthodon, 63, 70, 71 (Text-figs. 3 a-b), 72, 77 nilotica, Turraea, 128 nimbifera, Symphoristis, 55 niveocervina, Oxymachaeris, 43 nodiflorum, Solanum, 133 Nothris, 22 nubeculosa, Odites, 54 10 Annals of the Transvaal Museum nubica, Tephrosia, 128 nubigenum, Mesembrianthemum, 125 nudiuscula, Roripa, 126 nudus, Pelmatorycter, 91, 9a Numida, 2 Nyctinomus, 4 obelacma, Leucoptera, 41 obovata, Crassula natans, 126 Pleiospora, 127 obscurum, Panicum, 121 obsepta, Dragmatucha, 24 obtusifolia, Hippocratea, 130 Turraea, 128 ochraula, Pycnostola, 14 ochreolum, Acanthodon, 76 ochrozona, Orneodes, 35 octophora, Brachmia, 25 Odina, 130 Odites, 54 OECOPHORIDAE, 29, 53 Oinophila, 57 OLACACEAE, 125 olaefolius, Loranthus, 124 Oldenlandia, 133 OLEACEAE, 132 ominosa, Fumea, 45 Ommannei, Senecio, 135 Onebala, 25 onychotis, Crobylophora, 41 Opogona, 56 Opostega, 41 Opsigenes, 30 opsonoma, Eucosma, 48 orangana, Habenaria, 123 ORCHIDACEAE, 123 Orneodes, 34, 55 ORNEODIDAE, 34, 55 orthobathra, Depressaria, 31 Otomys, 1 14 ovalifolia, Gnidia, 13 1 ovata, Brachycorythis, 123 Plantanthera, 123 Oxymachaeris, 42 oxymoris, Oinophila, 57 oxyriaefolius, Senecio, 135 Pachycarpus, 132 pachydesma, Cnephasia, 9 pachypoda, Brassica, 126 Pachystigma, 134 palki, Georychus, 5 pallida, Eriochrysis, 121 pallidum, Galeosoma, 79 palliolatum, Thesium, 125 palmata, Ipomoea, 132 Paltodora, 13 Pancratium, 122 panduraefolius, Senecio, 135 paniculata, Phylica, 130 Pleiospora, 127 paniculatus, Monactinocephalus, 134 Panicum, 121, 122 pantomima, Microcolona, 53 PAPAVERACEAE, 126 papillosa, Eulophia, 124 Numida, 2, 3 paracma, Polyhymno, 19 Parapsectris, 17 parastacta, Opsigenes, 30 Paraxerus, 112 pardalina, Paromostola, 84 Parectopa, 39 Paromostola, 84 parviflora, Davalia transvaalensis, 132 parvus, Pelmatorycter, 108 PASSIFLORACEAE, 13 1 patersonae, Stasimopus, 95 Pavetta, 134 Pearsonia, 119, 127 pectinata, Crassula, 126 Schizaea, 120 pectiniformis, Pellaea, 121 pectinipalpis, Acanthodon, 73, 76 PEDALIACEAE, 133 pedicata, Schiffermuelleria, 29 peduncularis , Fleurya, 124 pedunculata, Crabbea, 133 Giesekia pharnaceoides, 125 Pegolettia, 134 Pelargonium, 128 Pellaea, 121 Pelmatorycter, 63, 91, 108, 109 pendens, Phthorimaea, 18 Pennisetum, 122 pentadecandra, Giesekia, 125 pentandra, Pharnaceum, 125 pentasticta, Aulotropha, 32 Pentatricha, 134 pentrandrum, Trianthema, 125 Pentzia , 135 Peperomia, 124 peploides, Crassula, 126 percivali, Cloeotis, 112 perfoliata, Arabis, 126 pericentra, Hesperarcha, 38 personatus, Spiroctenus, 88, 90 Petersii, Ficus, 124 petiolatum, Elaphoglossum, 121 petricola, Aloe, 122 petrophilus, Platymops, 4, 5 Phaenohoffmannia, 127 phaeocephala, Tineola, 44 PHALONIADAE, 8 Pharnaceum, 125 Phaseolus, 11 photaula, Orneodes, 55 Photodotis, 15 Phrixosceles, 39 Phthorimaea, 18, 19 Phyciodyta, 58 Phylica, 130 Index to Volume VI ii Phyllanthus, 130 PHYTOLACCACEAE, 125 Pienaarii, Aloe, 122 pilifera, Matricaria, 135 pilosa, Acalypha glabrata, 130 pilosum, Galeosoma, 79 pilulifera, Euphorbia, 129 pinifolia, Gnidia, 13 1 Vernonia, 134 pinnata, Merremia, 132 pinnatifida, Conyza, 134 pinnatifidum, Ceterach cordatum, 120 pinnulatus, Senecio, 135 PIPERACEAE, 124 planiscutatum, Galeosoma, 63, 78, 79 (Text-fig. 4 d), 80 Platanthera , 123 Platybathra, 38 Platymops, 4 platyneuron, Asplenium, 120 Platyptilia, 7 Plectranthus, 133 Pleiospora, 127 plicatus, Pachycarpus, 132 Plukenetia, 130 Plumbago, 41 plumosa, Eragrostis tenella, 122 plumosus, Trachypogon, 121 pluridentatum, Idiothele, 63, 101 (Text-fig. 12 b) pluripunctatum, Galeosoma, 63, 81; PL IV, fig. d podalyriaefolia, Pearsonia, 127 Podocarpus, 121 polita, Davalia, 132 Pollinia, 121 polyacantha, Berkheya, 135 polycephalus, Arthrosolen, 13 1 Polychrosis, 11 Polygala, 129 POLYGALACEAE, 129 Polyhymno, 19 polyphylla, Habenaria, 123 polypodantha, Habenaria, 123 POLYPODIACEAE, 120 polypodioides, Polypodium, 121 Polypodium, 120, 121 polystachya, Lessertia, 128 Lessertia perennans, 128 polystacta, Scardia, 47 Polystichum, 120 Porthmologa, 53 praecipua, Bucculatrix, 42 praemorsum, Asplenium, 120 praestricta, Parectopa, 39 Prays, 56 pretoriensis, Aloe, 122 Gladiolus, 123 primulina, Vernonia, 134 princeps, Dissotis, 13 1 Proceleustis, 54 Proctopus, 1 18 prolifera, Dryopteris, 120 proliferumy Polypodium , 120 Promalactis, 29 propinqua, Pearsonia, 127 propitia, Argyroploce, 12 prostrata, Melhania, 13 1 Proterochyta, 56 Protomacha, 31 pseudoglandulifera, Jatropha, 129 Psiadia, 134 Psilotrichum, 125 Psoralea, 128 Pteridium, 121 Pterinochilus, 63, 97, 98, 100, 102, 104 Pteris, 121 pteroides, Pellaea, 121 PTEROPHORIDAE, 7 ptychospila, Symphoristis, 55, 56 pubescens, Brachycorythis, 123 pubiflorum, Schizoglossum robustum, 132 Pueraria, 128 pulcherrima, Heeria, 130 pulchra, Disa, 124 Lotononis, 127 pulverulentus, Mungos, 114 punctata, Acalypha, 130 Sesbania, 128 punctatus, Andropogon inter medius, 121 Spiroctenus, 90 punctulatum, Blechnum, 120 pungens, Aspidium aculeatum, 120 Polystichum, 120 Pupalia, 125 purcelli, Damarchodes, 109 purpurascens, Ischaemum, 121 purpureo-auratus, Gladiolus, 123 purpureo-sericeum, Sorghum, 121 purpureo-sericeus, Andropogon , 121 pusilla, Eugenia, 13 1 pusillumy Trichomanes, 120 pycnocephalus, Carduus, 135 Pycnodytis, 15 Pycnostola, 13 pygmaea, Pachystigma, 134 Vangueria, 134 pyramidota, Acrocercops, 39 quadriaurita , Pteris, 121 quadripinnata, Pellaea, 121 quinquenervis, Loranthus, 124 quinquepartita,Trochomeria Hookeri, 134 racemosa, Gymnogramma, 38 radiatum, Lasiospermum, 135 Randii, Vernonia, 134 RANUNCULACEAE, 125 Ranunculus, 126 Raphanocarpus, 134 Raphionacme, 132 12 Annals of the Transvaal Museum rarum , Hymenophyllum, 120 rasum, Thesium impeditum, 125 Reckii, Holothrix, 123 Recurvaria, 20 regularis, Brachiaria, 122 Rehmanni, Eulophia, 124 Habenaria, 123 Poly gala, 129 Schizochilus, 123 rehmanniana, Acacia, 127 Rehmannii, Berkheyopsis, 135 Hygrophila 133 Lissochilus, 124 reniforme, Pelargonium, 128 repens, Dichondra, 132 Serpicula, 132 resedoides, Thesium, 125 resiniflua, Gardenia, 133 resonans, Ceromitia, 47 reticens, Stenoma, 34 revoluta, Gelechia, 17 RHAMNACEAE, 130 Rhinacanthus, 133 Rhinolophus, 112 rhomboideus, Senecio, 135 Rhus, 130 Rhynchosia, 128 rhytidophylla, Grewia, 130 rigens, Polygala, 129 rigida, Maerua, 126 rigidis, Rubus, 126 robertsi, Galeosoma, 82; PL IV, fig. b Rogersiae, Anisotes, 133 Rogersii, Acalypha punctata, 130 Cyphia, 134 Dyschoriste, 133 Lotononis, 127 Pearsonia, 127 Phyllanthus, 130 Pueraria, 128 Tragia, 130 Roripa, 126 ROSACEAE, 126 rostrata , Melhania, 13 1 rotundifolia, Dombeya, 13 1 RUBIACEAE, 133 rubromarginatus, Loranthus, 124 Rubus, 126 ruddi, Mungos, 114 Rudolphi, Vitex mooiensis, 133 rufa, Hyparrhenia, 121 rufescens, Indigofera, 128 rufus, Cymbopogon , 121 rugosella, Hapsifera, 59 Ruprechtii, Cymbopogon, 121 Hyparrhenia, 121 RUTACEAE, 128 rutaefolium, Asplenium, 120 Saccharum, 121 salicoma, Melasina, 45 salsuginosa, Ammannia, 13 1 Salvia, 133 Sandersonia, 122 sanguinea, Euphorbia, 129 Sanguisorba, 126 Sanseviera, 122 SANTALACEAE, 125 Sapheneutis, 45 sarmentosa, Kniphofia, 122 Satyrium, 123 Sauromys, 5 saxicola, Schistostephium, 135 scaberrimus, Corrallocarpus sphaero- carpus, 134 Scabiosa, 134 scabiosoides, Thelesperma, 135 scalmotoma, Promalactis, 29 scandicinum, Athyrium, 120 Scardia, 47 Schiffermuelleria, 29 schinziana, Ficus, 124 Schinzii, Acalypha, 129 Andropogon, 121 Crotalaria, 127 Crotalaria pilulicarpa, 127 Schistostephium, 135 Schizachyrium, 121 Schizaea, 120 SCHIZAEACEAE, 120 Schizochilus, 123 Schizoglossum, 132 Schlechteri, Adenia, 13 1 Crotalaria spinosa, 127 Vernonia, 134 Schlechteriana, Brachycorythis, 123 schonlandi, Pterinochilus, 97 Stasimopus, 93, 94, 95 schreineri, Acanthodon, 111 Galeosoma, 81; PI. IV, fig. c Scirpus, 122 Sclerophricta, 46 Scolopia, 13 1 SCROPHULARIACEAE, 133 SCYTHRIDAE, 37 Scythris, 37 seabrae, Cistugo, 112, 113 secularis, Eucryptogona, 45 secunda, Rhynchosia, 128 segetalis, Acalypha, 130 segmentata, Lopholaena, 135 Segregara, m Selaginella, 121 SELAGINELLACEAE, 121 semiberbe, Schizachyrium, 121 semiberbis, Andropogon hirtiflorus, 121 semnophanes, Gracilaria, 40 Senecio, 135 senegalensis, Hexalobus, 126 Pegolettia, 134 senensis, Acalypha, 130 sericea, Eulalia, 121 Index to Volume VI 13 sericea, Lessertia perennans, 128 sericocephala, Arthrosolen, 13 1 sericoflora, Lotononis, 127 Serpicula, 132 serrata, Brachiaria, 122 Worms kioldia, 13 1 serratum, Panicum, 122 Sesamothamnus, 133 Sesamum, 133 Sesbania, 128 sessiliflora, Aloe, 122 sessiliflorus, Anisotes, 133 sessilifolia, Lotononis , 127 Pearsonia, 127 seticoxa, Moggridgea, 107 sexangularis, Kalanchoe, 126 silberbauer, Dryodromas fulvicapilla, 1 17 silberbaueri, Otomys, 114 silvatica, Dryopteris, 120 Simaethis, 55 Simii, Aloe, 122 similis, Lasiosiphon, 13 1 sinuato-dentata, Ipomoea bathycolpos, 132 Sisyranthus, 132 Sisyroxena, 35 SOLANACEAE, 133 Solanum, 133 SOLIFUGAE, 63, 64 solitudinis, Lotononis, 127 Sonderi, Tragia, 130 Sorghum, 121 sparsisora, Hypolepis, 121 spartifoliella, Leucoptera, 41 spartioides, Crotalaria, 127 Thesium, 125 speciosa, Gerbera, 135 speluncae, Davallia, 120 Microlepia, 120 Spermacoce, 134 sphaerocarpus, Corrallocarpus, 134 Sphenandra, 133 spheroideum, Galeosoma coronatum, 63, 82; PI. IV, fig. ) yy ,, Barberton . 274 25 29 18 50 * J 5 y mitrata, Boror ... ... .. . 268 26 25 — 30 5? papillosa transvaalensis , Pretoria . 280 25 30 19 50 ? 9 ,, ,, Rustenburg. 280 22.5 29 18.5 50 y y ,, Rustenburg. .. 280 23 29 18 53 y y 5? damarensis , Windhuck ... .. . 280 22 16 11.5 38 yy y y •n at . 285 24 17 10 37 9 9 9 y ,, 9» . 293 24 15 9 37 3 GUTTERA EDOUARDI SYMONSI subsp. nov. Differs from the typical G. edouardi (Hartl.) from the coast of Natal in having only the faintest traces of chestnut coloration in the feathers of the lower neck, back, scapulars, chest and flanks and the white spots more distinctly blue, the whole general effect produced by these differ- ences being very marked. G. lividicollis Ghigi (Mem. Acc. Sc.Instit., Bologna, vi, pi. ii, 1905) does not appear to differ from the typical G. edouardi in the feather coloration, but has been founded upon the colora- tion of the skin of the neck. No notes were made upon the colour of the soft parts of this new subspecies at the time of their capture, and I am therefore unable for the present to detail them ; but Mr. R. E. Symons, who was kind enough to send a series of five specimens' to the Transvaal Museum, but is away from home at the time of writing, has promised to remedy the omission when next he is able to secure speci- mens ; the lack of these particulars does not, however, affect the status of the new subspecies. The specimens examined are two pairs of adults and an immature male, the last still retaining black and buffish down on the back of the head and short black feathers, some tipped with dirty white, on the throat ; the crest is also not fully developed. Essentially the immature specimen does not differ from the adults — but traces of bars instead of spots of bluish white are to be seen in the outer wing- coverts, abdominal region and flanks, and particularly on the tail feathers. The type series (of which T.M. No. 11585 is the type) was taken at Karkloof (3,500ft.), Natal, by Mr. R. E. Symons, after whom 1 have pleasure in naming the subspecies. Dimensions: Adults, length of wing from tip of primaries 260 275 mm ; tail 140-160 ; metatarsus 80 ; culmen 22-25. Immature, wing 950, tail 130, metatarsus 72, culmen 22. DESCRIPTIONS OF TWO NEW SPECIES OF A NEW SUBGENUS OF BATS AND A NEW SPECIES OF MOLERATS. By Austin Roberts. PLATYMOPS PETROPHILUS spec. nov. This bat has a superficial resemblance to Nyctinomus aegyptiacus (S. African specimens), but upon close examination is found to differ in having the ears more widely separated at the inner base ; the antitra- gus represented by a mere thickening of the lobe ; the horny points on the upper part of the ear commencing near the rim of the ear and not far down as in aegyptiacus ; the thumb very thick ;the first and fifth toes also very thick — equal to double that of either of the other three — and thickly covered above and on the outer surface with fine white bristly hairs, but with hardly any of the very long bristles found in aegyptiacus ; the ears are haired as in aegyptiacus , but the area around the fibulae is more sparsely covered with white hairs. The ears arise about 4mm. apart to a height of a little more than 1 mm. and the upper margin thence for about 13 mm. is practically horizontal, thence curving down- wards a little way, increasing the breadth of the ear to 15 mm., then downwards and very slightly inwards for about 9.5 mm. to a point whence an increasing tendency inwards for 4 mm. carries the lower margin in an almost straight line to the outer base ; the antitragus is represented by a broadening and thickening of the lobe, flattened or slightly folded on the upper surface, with only a slight emargination beyond its outer edge ; the tragus is like that of aegyptiacus, quadrate, rather narrower at the base than the tip, with a slight nitch in the outer edge. The horny points on the upper margin of the ear are five in number just below the horizon, the rest on the edge of the ear, pro- ducing a saw-like horizon. The lips are wrinkled, the nostrils sub-lateral and no glands appear to be present. The wings arise on the tibia at about one-third of the length of the tibia from the feet. The skull of the type is, unfortunately, fragmentary, having been crushed when the rock was overturned under which it had taken up its abode. Nevertheless, from what there is of it, the dentition is clearly seen to be numerically identical with that of N. aegyptiacus , but the skull is flattened as in Platymops Thos. (Ann. & Mag. N.H., ser. 7, vol. 17, p. 499, 1906) ;the premaxillae do not meet anteriorly, and the incisors are large and convergent : the foremost premclar (P2 ) is small, but easily seen without the aid of a lens, and is situated between the C °nd P4, well within the tooth row ; the lower anterior premolar is slightly smaller than the posterior premolar. Dimensions (taken from a spirit specimen) : Length of the head and body 53 mm., of the tail 38, of the free portion of the tail 15; diametre of the ear, vertically 14, horizontally 15; height of tragus 2.5; length of snout to the inner base of the ear 7 ; forearm 42 ; thumb 7 ; third digit, metacarpal 43, 1st phalanx 16, 2nd without the cartilage 15, with the Annals of the Transvaal Museum. 5 cartilage 22 ; fourth digit, metacarpal 41, 1st phalanx 15, 2nd phalanx without cartilage 8.5, with cartilage 11.5 ; fifth digit, metacarpal 27, 1st phalanx 11, 2nd with the cartilage 7.5; tibia 9; hind foot length (c.u.) 8, breadth 4 ; breadth of ankle pad 2. Skull : Length of dental series, from front of C to back of M3, 7.4; greatest length of mandible 13.5. Type: Adult male, in spirits, taken by G. van Dam at Bleskop, near Rustenburg, 2nd February, 1917. PLATYMOPS HAAGNERI spec. nov. A small species having an affinity to P. petrophilus in the dental formula, flattened skull, absence of a definite antitragus, shortness and number of the bristles on the toes, and in the horny points on the upper part of the ears being situated near the horizon of the ear ; but differing therefrom in its smaller size, the horny points on the ears only two in number ; the lower P2 much smaller than the P4, which is higher, the tips of Mlt P4 and C forming an almost straight line at a slight angle rising forward ; the upper incisors are also comparatively shorter than in petrophilus , and a cingulum can be seen in P2 with the aid of a lens. The lower incisors are bifid. The bony palate extends well behind the molars. Dimensions of the type in spirits: Length of the head and body 50 mm.; of the tail 32, of the free portion of the tail 13.5; ear, height 10, breadth 13.5; height of the tragus 2.3; length of forearm 37; pollex (c.u.) 5; third digit, metacarpal 37, 1st phalanx 14, 2nd without carti- lage 12; fourth digit, metacarpal 36, 1st phalanx 12.5, 2nd phalanx without cartilage 6; fifth digit, metacarpal 25, 1st phalanx 10, 2nd 4.7; tibia 10; hind foot, length (c.u.) 6.5 Skull: Greatest basal length 16.2; zygomatic width 10 ; width at mastoid 9.8 ; width of brain case 9 ; in- terorbital constriction 3.7; basilar length 13.1; palate length 5.5; length of dental series, from front of C to back of M3, 6 ; width of palate taken outside the molars 7 ; greatest length of mandible 12. Type: Adult male from Keetmanshoop, Damaraland, presented to the Museum by Mr. Sigmund Haagner. These two species apparently belong to the genus Platymops Thos., but differing as they do in the dental formula, the P2 being present, I propose to place them in a new subgenus bearing the name of SAUROMYS, of which P.haagneri maybe taken as the genotype. The fol- lowing figures will indicate the difference in the height of the skull of N. aegyptiacus and the present new species, the measurements having been taken from the tips of the P4 vertically parallel to the top of the premaxilla : N. aegyptiacus 6.5 mm., P. petrophilus 4.2, P. haagneri 4. GEORYCHUS PALKI spec. nov. Most closely allied to G. anomalus mihi in having the nasals long and narrow and the same buffish grey coloration ; but differing from this and all other members of this group in having a very distinct semi- circular notch in the upper anterior angle of the ocular area of the skull, this notch effecting a thinning of the arch of the maxilla above the antorbital foramen and opposite the antorbital process. In two adult specimens the hindmost molar is very large, equal in diametre to the foremost tooth. Dimensions: Length of the head and body 150 mm.; tail 22; hind 6 Annals of the Transvaal Museum. ,foot (s.u.) 29. Skull: Extreme length from the tips of the incisors 42.4; length from the apex of the premaxilla 38.7, from the tips of the nasals 37 ; basilar length 32.5 ; greatest zygomatic width 29.2 ; width of brain case 16.8; mastoid width 20; inter-orbital constriction 8.4; width of premaxilla 8.7: width of incisors at the exposed base 6.4; length of molar series 7.5; diastema 12.8; length of bony palate 24.2; greatest diametre of bullae, including the internal projection, 10.3 ; nasals 15.6 x 4.1. Type: Old male, No. T.M. m 2085, taken at Venterskroon, on the banks of the Vaal River, Potchefstroom District, 24th March, 1917. Also a series of five younger specimens from the same colony, showing the same characters ; but unfortunately the only female captured is not mature and the number of mammae is not known. I have named this animal after Mr. Robert Palk, who materially assisted me in effecting their captvre. Besides these specimens, two others were captured in another colony, an adult female and a young adult male, which have all the characters, including the number of mammae, of G. natalensis mihi, though they are somewhat paler on the muzzle. This again illustrates the necessity for carefully noting which colonies specimens were cap- tured in, each species apparently living upon a particular vegetable food. ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM VOLUME VI PART 2 containing Descriptions of South African Micro-Lepidoptera By E. Meyrick, B.A., F.R.S. A New Species of Warburgia from the Transvaal By Mrs R. Pott PRINTED AT THE UNIVERSITY PRESS CAMBRIDGE, ENGLAND 1918 ANNALS MEDEDELING OF THE VAN HET TRANSVAAL MUSEUM IAR 28 1919 ^<9 * ' . Jj/0nal Vol. 6 Part 2 DESCRIPTIONS OF SOUTH AFRICAN MICRO-LEPIDOPTERA By E. Meyrick, B.A., F.R.S. PTEROPHORIDAE Platyptilia bullifera n.sp. $ 20 mm. Head pale brownish, face and sides of crown whitish. Palpi 2J , brownish mixed with dark fuscous, base and apex white. Thorax pale brownish, posteriorly whitish. Abdomen whitish, on sides and posteriorly mixed with fuscous. Forewings cleft to beyond f , second segment narrow at base and posteriorly dilated, apex pointed, termen of first segment concave, oblique, of second concave, little oblique; pale brownish suffusedly irrorated with white except towards anterior half of costa, towards costa suffused with reddish-fuscous, costal edge suffused with dark fuscous towards middle; a trapezoidal dark reddish-fuscous spot crossing middle of first segment, narrowed downwards, edged anteriorly with white suffusion broader towards costa, and posteriorly by a white line ; a trapezoidal dark reddish-fuscous spot occupying basal half of second segment; both segments beyond these markings light red-brownish sprinkled with white: cilia whitish, on costal spot dark fuscous, on termen with basal third brown edged by a dark fuscous line, twice interrupted with white on first segment, within cleft grey mixed with whitish on lower margin of first segment, on dorsum with three dark grey patches tipped with black, viz. one moderate at f of wing, one narrow beneath cleft, and one broad from near this to tornus. Hindwings dark grey, third segment f of first; cilia grey, tinged with whitish on lower angle of second segment, on dorsum with an inconspicuous tooth of three or four blackish-tipped scales at f of third segment, and three or four scattered similar scales near before this. Zululand, Eshowe, in January (Janse); one specimen. A very distinct species, of early type. 1 8 Annals of the Transvaal Museum CARPOSINIDAE Carposina exsanguis n.sp. $ 15 mm. Head and thorax white, patagia faintly tinged with ochreous. Palpi ochreous-whitish finely sprinkled with dark fuscous. Forewings elongate, rather narrow, posteriorly slightly dilated, costa gently arched, apex pointed, termen nearly straight, oblique; ochreous-whitish, faintly clouded with very pale greyish-ochreous ; an indistinct mark of fuscous suffusion along basal sixth of costa, and five very small faint fuscous or greyish-ochreous costal spots between this and apex : cilia ochreous-whitish faintly speckled with light grey. Hindwings and cilia whitish. Transvaal, Barberton, in December (Janse); one specimen. PHALONIADAE Euxanthis umbraculata n.sp. $ 16-19 mm. Head ochreous-whitish. Palpi 3, whitish-ochreous. Thorax whitish, partially suffused with pale ferruginous. Abdomen dark grey, apex ochreous-whitish. Forewings elongate, rather narrow, somewhat dilated posteriorly, costa hardly arched, faintly sinuate in middle, apex obtuse, termen straight, oblique; light ferruginous-ochreous, more or less mixed with whitish; some ferruginous-brown marking on basal area, and anterior half of costa suffused with ferruginous-brown; a rather narrow ferruginous- brown median fascia, mixed with dark fuscous on dorsal half, obtusely angu- lated or tending to be interrupted above middle; a small dark fuscous discal dot beyond this ; three small brown spots on costa between this and terminal fascia; a suffused ferruginous-brown terminal fascia, tending to be produced anteriorly in suffused streaks along veins towards cell : cilia whitish-ferruginous, barred or sometimes generally suffused with rather dark grey, with dark grey subbasal line. Hindwings dark grey; cilia whitish, with grey subbasal line. Transvaal, v.d. Merwe and Middelburg, in October and December (Janse) ; two specimens. TORTRICIDAE Tortrix corroborata n.sp. 15 mm. Head and thorax pale brownish-ochreous. Palpi 3^, brownish- ochreous. Antennal ciliations 1. Abdomen dark grey, anal tuft ochreous- whitish. Forewings suboblong, moderate, costa anteriorly strongly, posteriorly hardly arched, without fold, apex obtuse, termen faintly sinuate, hardly oblique; pale brownish-ochreous, tinged with silvery-grey on dorsal half; basal patch little marked, partially suffused with light ferruginous-brown except towards costa, edge irregularly curved; central fascia moderate, oblique, ferruginous-brown, on dorsal half dilated and suffused with grey posteriorly except on edge, costal edge dark fuscous; costal patch semioval, ferruginous-brown, suffused with blackish on costal edge, connected by an irregular stria with termen above tornus : cilia pale ochreous, basally somewhat dotted or barred with blackish, more strongly above apex and beneath tornus. Hindwings dark grey; cilia pale greyish, with dark grey subbasal shade. Zululand, Nkandhla Forest, in January (Janse); one specimen. Tortrix metapyrrha n.sp. £ 18 mm. Head and thorax whitish-yellow. Palpi 2^, pale yellowish. Abdomen pale whitish-ochreous. Forewings elongate, rather narrow, costa Annals of the Transvaal Museum 9 slightly arched, apex obtuse-pointed, termen almost straight, rather oblique; pale yellowish ; a faint greyish dot on lower angle of cell : cilia light ochreous- fulvous, towards tornus yellow- whitish. Hindwings and cilia ochreous-whitish. Transvaal, Pretoria, in September (Janse); one specimen. Cnephasia pachydesma n.sp. $ io mm. Head whitish-ochreous, centrally tinged with grey. Palpi i|, whitish-ochreous. Thorax whitish -ochreous-grey. Abdomen grey. Forewings elongate, moderate, posteriorly somewhat dilated, costa gently arched, apex obtuse, termen slightly rounded, oblique; pale ochreous; markings dark grey mixed with blackish; basal patch occupying J of wing, edge irregular, angulated below middle; central fascia moderately broad, straight, oblique; an almost terminal fascia, broad on costa, narrowed downwards to tornus; discal area between these markings broadly suffused with grey sprinkled with blackish: cilia pale ochreous, beneath tornus dark grey. Hindwings dark grey; cilia grey, with dark grey subbasal shade. Zululand, Nkandhla Forest, in January (Janse); one specimen. Cnephasia flavisecta n.sp. $ io mm. Head ochreous-yellow, centrally yellow- whitish. Palpi 2, ochreous-yellowish. Thorax ochreous-yellowish, patagia grey except shoulder. Abdomen grey. Forewings elongate, posteriorly somewhat dilated, costa moderately arched, apex obtuse, termen nearly straight, oblique; golden- yellowish ; markings slaty-grey, irregularly sprinkled with black, and partially mixed with whitish ; a moderate basal patch, outer edge hardly curved, vertical, including an elongate blotch of ground-colour from base of costa and a small spot below fold ; a moderate direct median fascia, somewhat expanded towards dorsum, its posterior edge suffused with black on upper half; a narrow fascia from f of costa to tornus, and one just before apex marked with black on edges, these connected by a slender bar in disc: cilia whitish-yellowish. Hind- wings pale grey; cilia grey-whitish. Zululand, Eshowe, in January (Janse); one specimen. EUCOSMIDAE Eucosma inscita Meyr. $ ii— 12. mm. Head and thorax grey, thorax sometimes with blackish median transverse bar. Palpi fuscous irrorated with whitish. Abdomen rather dark fuscous. Forewings elongate, posteriorly slightly dilated, costa gently arched, apex obtuse, termen faintly sinuate, somewhat oblique; brownish-grey; costa marked with alternate groups of very fine whitish and blackish strigulae ; an angulated median transverse fascia composed of several confluent leaden striae ; short leaden marks rising from costal groups of whitish strigulae beyond this ; ocellus reaching § across wing, laterally margined with leaden-metallic streaks of which posterior is cut in middle by a very fine black dash from its interior, three fine black linear marks within ocellus above this, and some irregular short confused black dashes above and before its upper part; apex dark fuscous: cilia grey sprinkled with white. Hindwings with 3 and 4 connate; dark grey, lighter anteriorly; cilia light grey, with darker subbasal shade. Natal, Umkomaas, in January (Janse) ; four specimens. I have redescribed this obscure species from this improved material, as its distinguishing features were not previously adequately expressed. I — ;2 10 Annals of the Transvaal Museum Eucosma bisecta n.sp. ^ ii mm. Head and palpi dark slaty-grey, face blackish. Thorax dark fuscous, posterior half suffusedly mixed with ochreous. Abdomen dark grey. Posterior tibiae tufted with rough projecting scales beneath. Forewings elongate, posteriorly rather dilated, costa gently arched, apex rounded, termen somewhat rounded, oblique, sinuate-indented above middle; blackish-fuscous; dorsal half from base to beyond middle marked with irregular transverse blue-leaden strigae partially irrorated with whitish-ochreous ; basal area suffused with greyish- violet above this; two greyish- violet streaks from about § of costa to beyond middle of disc, thence acutely angulated to near dorsum, touched with white on costa ; a similar oblique streak from f of costa, angulated in middle to tomus, followed on upper half by a fine line of pale ochreous- yellowish irroration, its lower portion edged anteriorly with pale ochreous- yellowish and followed by an irregular pale ochreous-yellowish patch repre- senting ocellus and including two or three blackish dots; two conspicuous white direct strigulae on costa at f, terminated beneath by short oblique ochreous-yellowish marks, beyond which is a greyish-violet dot : cilia grey, round apex with blackish basal and shorter apical lines, whitish-suffused at both extremities of apical line, lower forming a whitish spot extending to basal line, beneath this with a spot of dark fuscous suffusion. Hindwings with 3 and 4 stalked, 6 and 7 stalked; dark grey, towards base with a hyaline space beneath cell; cilia light grey, with dark grey subbasal shade. Zululand, Eshowe, in January (Janse); one specimen. Eucosma desipiens n.sp. cJ 11 mm. Head, thorax, and abdomen grey-whitish. Palpi densely rough- scaled, white, with oblique light grey median band. Forewings elongate, rather narrow, costa gently arched, with moderate fold from base to f , apex obtuse-pointed, termen faintly sinuate, oblique; light grey with tips of scales white; costa from beyond fold marked with dark fuscous strigulae, and three small dark fuscous spots; small scattered undefined spots of fuscous suffusion towards dorsum and in disc posteriorly, and an indistinct streak of fuscous suffusion from beyond middle of costa to tornus, angularly projecting in middle, with a line of blackish scales in disc running into angle; apical area of wing dark grey; ocellus margined laterally with thick silvery-whitish streaks, and containing three short black linear marks : cilia whitish-grey sprinkled with white and fuscous, at apex with a fuscous bar. Hindwings with 3 and 4 stalked; pale grey; cilia grey- whitish. Transvaal, Alberts-mine, in January (Janse); one specimen. Eucosma glyphicodes n.sp. $ 20 mm. Head light brownish, face suffused with dark fuscous. Palpi dark fuscous, towards base white. Thorax blackish, patagia mixed with ochreous-brown and whitish. Abdomen fuscous. Forewings elongate, costa anteriorly gently arched, posteriorly nearly straight, apex obtuse, termen nearly straight, somewhat oblique; white, partially suffused with whitish- ochreouS; basal portion mixed with pale leaden-grey and irregularly marked with black, its margin irregularly black, rather oblique, enclosing a clear white spot on dorsum; between this and central fascia are two or three interrupted irregular transverse pale leaden-grey streaks, and two black dots on costa; central fascia little oblique, formed of irregular pale leaden-grey and black markings somewhat mixed with light reddish-ochreous, especially a black spot, on middle of costa and an irregular longitudinal black mark projecting pos- Annals of the Transvaal Museum ii teriorly in disc; two small black spots on costa beyond this; a leaden-grey patch resting on upper half of termen, including an irregular black spot, apex and upper half of termen marked with small black spots; ocellus limited anteriorly by a leaden-grey streak confluent with this : cilia grey, basal half barred with light greyish-ochreous, at tornus with an ochreous-whitish patch, at apex blackish-grey. Hindwings with 3 and 4 stalked; grey; cilia grey. Natal, Sarnia, in November (Mrs Curry) ; one specimen. Eucosma melanaula Meyr. Zululand, Eshowe, in January (Janse); an Indian species, not previously recorded from South Africa; the larva feeds on cultivated Leguminosae (Cajanus, Phaseolus) , and therefore doubtless the species has been artificially introduced. Polychrosis ephippias Meyr. Natal, Umkomaas, in January (Janse); another common Indian species, not previously noticed from South Africa; the larval habits are unknown. Argyroploce caryocoma n.sp. (J 13 mm. Head and palpi dark ferruginous-brown. Thorax white, dorsally and on shoulders mixed with dark fuscous, at apex of patagia with deep ferruginous, dorsal crest double, deep ferruginous. Abdomen grey-whitish, segmental margins grey. Forewings elongate, moderate, posteriorly rather dilated, costa gently arched, apex obtuse, termen somewhat rounded, hardly oblique; 7 and 8 very shortly stalked; whitish; costa marked with fine oblique blackish strigulae suffused with brown ; some grey suffusion at base and along dorsurn to near middle; raised black dots beneath fold near base and before middle; an ill-defined grey transverse blotch from of costa to fold, some pale leaden-grey marbling in disc beyond this; central fascia beyond middle, little oblique, narrow, very irregular, more or less interrupted in middle and towards dorsum, grey sprinkled with black and marked with black on posterior margin near costa, preceded on lower half by a pale leaden-grey striga ; a pale brownish streak beneath costa posteriorly, crossed by some oblique blue-leaden marking, beneath this two blackish dots ; ocellus edged anteriorly by a long pale bluish- leaden striga and posteriorly by a shorter silvery-whitish striga, brown mixed with black and white hairscales, closed above by a brown spot, terminal area beyond this brown with a white striga: cilia grey-whitish with two or three dark fuscous scales, towards tornus white. Hindwings with 3 and 4 short- stalked; grey; cilia grey-whitish, with grey subbasal shade. Natal, Umkomaas, in January (Janse); one specimen. Argyroploce erythropa n.sp. cJ 15 mm. Head reddish-fuscous. Palpi curved, ascending, dark reddish- fuscous, terminal joint very short, pale. Thorax fuscous. Abdomen grey. Forewings elongate, moderate, posteriorly rather dilated, costa slightly arched, apex rounded-obtuse, termen rounded, somewhat oblique; fuscous mixed with dark fuscous; costa minutely strigulated with pale irroration; two confluent dull red dots transversely placed on end of cell; a light brownish marginal streak round apex and upper part of termen, attenuated downwards to a point : cilia fuscous, finely whitish-sprinkled, with a dark fuscous subbasal line. Hindwings rather dark grey; cilia grey, becoming whitish towards tips, dorsal margin with an erect fringe of expanded whitish hairs. Hindwings beneath with veins 2-4 clothed with a patch of rough grey hairscales except towards termen. Natal, Umkomaas, in January (Janse); one specimen. 12 Annals of the Transvaal Museum Argyroploce aster ota n.sp. $ 1 6 mm. Head grey-whitish, forehead tinged with yellowish. Palpi light greyish sprinkled with white, second joint with two spots of dark fuscous irroration. Thorax grey-whitish, anteriorly suffused with darker grey, shoulders and crest tinged with yellowish. Abdomen dark grey. Forewings elongate- triangular, costa gently arched, apex obtuse, termen straight, somewhat oblique; dark grey, mixed here and there with olive-greenish and strewn with irregular silvery-leaden spots; costa suffusedly mixed with blackish, with pairs of direct indistinct whitish strigulae; an interrupted slender black longitudinal streak in disc from § to near termen ; a suffused white fascia from | of costa to dorsum before tornus, interrupted by grey suffusion round this streak ; apical area greenish-fulvous spotted with silvery-leaden : cilia grey sprinkled with white and dark fuscous. Hindwings rather dark grey; cilia whitish-grey, with dark grey subbasal shade. Zululand, Eshowe, in January (Janse); one specimen. A rgyroploce propitia n.sp. J ii mm. Head grey-whitish. Palpi whitish partially mixed with grey, second joint with long hairs beneath. Thorax whitish, anteriorly and posteriorly mixed with ochreous, shoulders with dark grey spots. Abdomen dark grey, anal tuft whitish towards tips. Forewings elongate, posteriorly dilated, costa gently arched, apex obtuse, termen nearly straight, somewhat oblique; whitish, somewhat sprinkled with grey, anterior half of costa marked with dark grey strigulae, posterior half dark fuscous with five pairs of white strigulae; basal patch somewhat strigulated with dark red-brown on dorsal half, its outer edge marked by a blackish stria interrupted beneath costa and slightly bent in middle; central fascia somewhat oblique, on upper half narrow, blackish-grey, on lower moderately broad, grey suffused with pale ochreous, posterior edge angulated in middle; beyond this a rather narrow irregular fascia of leaden-grey suffusion ; anterior edge of ocellus formed by a thick pale leaden-grey streak; a roundish greyish-ochreous patch towards apex, extending indistinctly downwards into ocellus, in which is a short blackish mark, and connected with middle of termen by a short oblique suffused dark grey streak ; a small dark red-brown apical spot ; cilia dark slaty- fuscous, with dark red-brown basal line, and red-brown tornal patch. Hind- wings with 3 and 4 short-stalked ; blackish -grey, with a subdorsal fold white- scaled internally and containing a pencil of long grey hairs; cilia grey, with dark grey basal line. Zululand, Eshowe, in January (Janse); one specimen. Laspeyresia victrix n.sp. 10 mm. Head and thorax dark fuscous, irrorated with whitish. Palpi white somewhat sprinkled with dark fuscous. Abdomen dark grey. Forewings elongate-triangular, costa hardly arched, apex obtuse, termen bowed, some- what oblique; blackish, with tips of scales white, forming a fine nearly regular transverse striation; a rosy-leaden-metallic transverse streak at hardly reaching costa or dorsum; a thicker rosy-leaden-metallic streak from middle of costa to beyond middle of dorsum, thinner and tinged with pale blue towards costa; space between this and next markings wholly blackish except on a patch below middle; three short bluish-leaden strigae from costa posteriorly ; ocellus margined anteriorly by a thick rosy-leaden-metallic streak reaching more than half across wing, posteriorly by a short and slender bluish- Annals of the Transvaal Museum i3 leaden-metallic streak, and crossed by four black lines; a terminal streak of whitish irroration : cilia light grey, with black basal line, iridescent on outer half, irrorated with black towards tips round apex. Hindwings with 3 and 4 connate; fuscous, paler and thinly scaled towards base, suffused with dark fuscous posteriorly; cilia light grey, with dark fuscous subbasal shade. Natal, Umkomaas, in January (Janse); one specimen. Laspeyresia areata n.sp. cJ 13 mm. Head dark fuscous irrorated with white. Palpi white, irregularly sprinkled with dark fuscous. Thorax dark fuscous, partially sprinkled with white, patagia with white apical patch. Abdomen grey. Forewings elongate- triangular, costa anteriorly straight, posteriorly gently arched, apex obtuse, termen bowed, somewhat oblique; blackish; basal fourth strewn with suffused white strigulae; costa beyond this with about eight pairs of white strigulae whence rise indistinct short fine blue-leaden strigae; a large white rather oblique transverse blotch resting on dorsum beyond middle and reaching more than half across wing, anterior edge convex, posterior confluent in disc with a broad patch of ochreous-white suffusion extending across wing from five posterior pairs of strigulae to tornus, including a silvery-metallic dorsal streak representing anterior margin of ocellus and a small mark on its posterior margin ; the narrow terminal streak beyond this is finely speckled with whitish : cilia grey. Hindwings with 3 and 4 connate; fuscous, lighter towards base, suffused with dark fuscous posteriorly; cilia ochreous-white, with dark grey basal shade. Transvaal, Pretoria, in November (Janse); one specimen. Laspeyresia excoriata n.sp. cJ 10 mm. Head and palpi whitish-grey-ochreous, crown spotted with light grey. Thorax whitish-grey-ochreous, spotted and anteriorly suffused with dark fuscous. Abdomen grey. Forewings elongate, posteriorly dilated, costa slightly arched, apex obtuse, termen fain tty sinuate, somewhat oblique; dark fuscous, from base to beyond middle irregularly and coarsely mottled with whitish-ochreous, beyond this with extreme tips of scales whitish-ochreous ; 'costa from near base to beyond middle with pairs of very fine whitish strigulae and beyond this with four single stronger white strigulae; a curved violet- leaden line running from last pair (beyond middle of costa) to anterior margin of ocellus, and an oblique line from second single strigula to a white mark on termen beneath apex; subcostal space between and beyond these yellow- ochreous, connected beneath with a yellow-ochreous patch representing ocellus and containing five longitudinal blackish marks, anterior limiting line broken and margined with ochreous-whitish towards lower extremity; a terminal streak of black and whitish speckling : cilia grey, towards tornus tinged with whitish. Hindwings with 3 and 4 stalked; grey, rather thinly scaled, veins darker; cilia whitish-grey, with dark grey basal line. Zululand, Nkwaleni, in January (Janse); one specimen. GELECHIADAE Pynocstola Meyr. All the South African species hitherto referred by me to Paltodora belong to this genus, which I have characterised as distinguished from Paltodora by the scaling of the palpi ( Ent . Mo. Mag. 1917, 113). i4 Annals of the Transvaal Museum Pycnostola crater aula n.sp. 27 mm. Head whitish mixed with grey. Palpi whitish, second joint irrorated with dark grey except apical edge, terminal joint anteriorly irrorated with dark grey. Thorax whitish, with a dorsal stripe of greyish suffusion, a fuscous line on each side of back, and a blackish blotch on shoulder. Abdomen fuscous, apex whitish. Forewings elongate, very narrow, costa faintly sinuate, posteriorly slightly arched, apex pointed, termen sinuate, very oblique; whitish; costa and veins suffused with blackish except on dorsal area beneath fold; some brown suffusion towards base of costa, above end of cell, and on a subterminal streak; large blackish dots suffused with brown beneath costa at and a short very fine black line along fold towards base; a blackish dot beneath fold at \ of wing; stigmata blackish, discal approximated, plical very obliquely before first discal ; a slender blackish streak along termen : cilia on costa white with three blackish bars, tips grey, on termen whitish with two blackish-grey shades, on tornus ochreous-grey-whitish. Hindwings grey; cilia light greyish-ochreous, tips whitish. Natal, Standerton, in February (Janse); one specimen. Pycnostola merista n.sp. $ 17 mm. Head white, centre of crown sometimes faintly fuscous-tinged. Palpi white, second joint irrorated with rather dark fuscous except apex, tuft short, terminal joint with median band of dark fuscous irroration. Thorax white irregularly mixed with rather dark fuscous. Abdomen whitish. Fore- wings elongate, narrow, costa gently arched, apex pointed, termen extremely obliquely rounded; white, with scattered dark fuscous scales; markings formed by dark fuscous irroration; a short mark from costa near base, and a spot on base of dorsum; narrow irregular oblique fasciae at | and plical stigma visible as a darker mark on second; discal stigmata represented by round spots at middle and f , first confluent with larger spots beyond it on costa and dorsum, second confluent with a spot on tornus ; an irregular trans- verse spot across wing towards apex, more or less confluent with tornal spot, and two small spots on costa before and beyond this : cilia whitish with a dark grey shade, round apex with a black basal line, and some black scales near base on terminal markings, towards tornus tinged with pale greyish**, ochreous. Hindwings light blue-grey; cilia whitish-ochreous tinged with grey, at apex with a small grey spot at tips. Transvaal, Pretoria, in September and October (Janse); two specimens. Pycnostola ochraula n.sp. cJ 16 mm. Head white, with a few pale grey specks. Palpi white, second joint grey except apex, tuft very long, terminal joint anteriorly with a few grey specks. Thorax white speckled with grey. Abdomen grey, anal tuft whitish. Forewings elongate, narrow, costa slightly arched, apex acutely pointed, termen faintly sinuate, extremely oblique; grey, suffusedly irrorated with white and sprinkled with dark grey; a fulvous-ochreous subcostal line from near base to f, with a blackish dot. beneath its extremity; a fulvous- ochreous line along fold throughout, marked with two black dots, the second being plical stigma; discal stigmata black, approximated, first very obliquely beyond plical, surrounded or almost connected with fulvous-ochreous; short fulvous-ochreous streaks above each of these, and a slender irregular fulvous- ochreous streak near and parallel to termen : cilia whitish-grey, with two faint dark lines. Hindwings light bluish-grey; cilia whitish-ochreous-grey. Annals of the Transvaal Museum 15 Transvaal, Pretoria, in September (Janse); one specimen. Nearly allied to illuminata, but distinguished by clear white head and suffusion of forewings. Photodotis spilodoma n.sp. $ 13 mm. Head whitish. Palpi whitish, second joint sprinkled with dark fuscous except at apex, terminal joint with subapical band of dark fuscous irroration. Antennae whitish ringed with dark fuscous, terminal § forming six dark fuscous bands separated by single whitish rings. Thorax whitish with a few dark fuscous scales. Abdomen ochreous-whitish sprinkled with fuscous. Forewings elongate, narrow, costa gently arched, apex obtuse, termen ex- tremely obliquely rounded; 3 present; dark fuscous irrorated with whitish, dorsal area on anterior half of wing suffused with whitish-ochreous, anteriorly extending § across wing; a blotch of blackish irroration representing plical and first discal stigmata, and a small spot on costa above it; a small brownish spot on tornus, connected by a cloudy blackish dot surmounted by a similar ochreous dot with indistinct dark second discal stigma ; a small cloudy whitish spot on costa at f ; two short ochreous streaks above and two below apex, separated by blackish irroration : cilia greyish, sprinkled with whitish and with base whitish on termen, with two darker grey shades. Hindwings grey; cilia light grey, base ochreous-whitish. Zululand, Eshowe, in January (Janse) ; one specimen. Vein 3 of forewings is absent in type of genus, but this species is in all respects so closely allied that the discrepancy should evidently be disregarded. Pycnodytis n.g. Head with appressed 'scales; ocelli small, posterior; tongue developed. Antennae f, in $ simple, basal joint elongate, without pecten. Labial palpi long, recurved, second joint thickened with dense scales rather loose beneath and triangularly expanded with hairs on apical half above, terminal joint shorter than second, slender, acute. Maxillary palpi very short, filiform, appressed to tongue. Posterior tibiae clothed with rough hairs above. Fore- wings with 2-4 parallel, 6 and 7 out of 8, 7 to costa, n from middle. Hind- wings under 1, narrow- trapezoidal, apex strongly produced, acute, termen emarginate, cilia 2; 2-5 remote, parallel, 6 and 7 approximated towards base. Allied to Aristotelia, but palpi much as in Trichotaphe. Pycnodytis erebaula n.sp. $ 11-12 mm. Head light grey, sidetufts tinged with brownish. Palpi dark fuscous, extreme apical edge of second joint white, terminal joint whitish, tip dark grey. Antennae dark grey, on apical half with four remote white dots. Thorax light grey more or less tinged with brownish. Abdomen pale grey. Forewings elongate-lanceolate, acute; dark fuscous, in one specimen with dorsal half suffused with brownish; stigmata black, plical obliquely before first discal, indistinct or obsolete: cilia grey, towards base suffusedly barred with dark fuscous. Hindwings and cilia light grey. Zululand, Eshowe, in January (Janse); two specimens. Aristotelia stipella Hlibn. cJ 2 5-6 mm. A small form with subbasal yellow blotch from dorsum broad and reaching nearly to costa, yellow spots beyond middle of dorsum and in disc at § well-developed, in one specimen the whitish costal spot pro- duced in an irregular streak to termen above tornus; I consider the examples i6 Annals of the Transvaal Museum undoubtedly referable to the southern form of this European species, not previously recorded from Africa but easily overlooked. Natal, Durban, in August and December (Janse); three specimens. Leuronoma n.g. Head smooth; ocelli moderate, far posterior; tongue developed. Antennae f, basal joint elongate, without pecten. Labial palpi very long, recurved, second joint with appressed scales, hardly thickened, terminal joint as long as second, slender, acute. Maxillary palpi very short, filiform, appressed to tongue. Posterior tibiae clothed with long fine rough hairs above. Forewings with 2 from towards angle, 7 and 8 stalked, 7 to costa, 11 from middle. Hind- wings 1, elongate-trapezoidal, termen slightly sinuate, cilia 1; 3-5 separate, rather approximated at base, 6 and 7 stalked. Type chlorotoma Meyr. Besides the two following species I refer to this genus zymotis Meyr., in which species the terminal joint of palpi is shorter than the second, the termen of hindwings more sinuate, cilia i|. The genus is somewhat intermediate between Telphusa and Aristotelia. Leuronoma chlorotoma n.sp. $ 16 mm. Head pale ochreous. Palpi whitish sprinkled with fuscous and dark fuscous. Thorax brownish mixed with grey, patagia suffused with pale ochreous, shoulders dark fuscous. Abdomen grey, apex whitish-ochreous. Forewings elongate, narrow, costa slightly arched, apex obtuse-pointed, termen extremely obliquely rounded ; rosy-brown, somewhat mixed irregularly with grey and towards dorsum suffused with grey, all veins except towards dorsum marked with slender rather irregular whitish-ochreous lines; plical stigma cloudy, blackish : cilia ochreous-whitish, towards base faintly rosy- tinged, with basal and antemedian lines of black points, and two grey posterior lines. Hindwings light slaty-grey; cilia whitish-grey. Transvaal, Pretoria, in November (Janse); one specimen. Leuronoma veter ascens n.sp. $ 11-12 mm. Head and thorax light brownish. Palpi light brownish sprinkled with dark fuscous, terminal joint as long as second. Abdomen pale ochreous irrorated with grey. Forewings elongate, narrow, costa gently arched, apex pointed, termen very obliquely rounded; light brownish, irregularly sprinkled with dark fuscous and some blackish scales; stigmata blackish, plical rather obliquely before first discal ; obscure opposite spots of dark fuscous irroration on costa and dorsum just beyond second discal: cilia light brownish, sprinkled with dark fuscous towards base. Hindwings grey; cilia light grey. Natal, Umkomaas and Stella Bush, in January (Janse); two specimens. Epithectis lacunosa n.sp. (J $ 10 mm. Head whitish irrorated with dark grey or blackish, sidetufts slightly ferruginous-tinged. Palpi grey irrorated with black, second joint with white spot below middle and extreme tip white, terminal joint white with two blackish bands. Thorax grey-whitish tinged with ferruginous, shoulders blackish-grey. Abdomen light grey, two basal segments whitish- ochreous, anal tuft of whitish. Forewings elongate, narrow, costa gently arched, apex pointed, termen extremely obliquely rounded; reddish-ochreous or ferruginous ; markings grey irrorated with black ; a narrow fascia from base of costa to dorsum before middle; a narrow oblique fascia from costa at J, Annals of the Transvaal Museum i7 below middle running into a narrow fascia which runs from a flattened- triangular blotch on middle of costa to dorsum beyond middle and coalesces there with first fascia; a patch of irregular marbling towards costa posteriorly, connected by a very irregular blotch with dorsum before tornus, edged posteriorly by a white mark near dorsum : cilia grey, on termen mixed with black towards base. Hindwings slaty-grey; cilia grey. Zcjluland, Nkwaleni, in January; Transvaal, Pretoria, in February (Janse); two specimens. Parapsectris ferulata n.sp. $ 13-14 mm. Head and thorax whitish irregularly sprinkled with dark fuscous. Palpi whitish, second and terminal joints each with base and supra - median band dark fuscous. Abdomen whitish-grey, second segment with a whitish-ochreous dorsal patch. Forewings elongate, narrow, costa gently arched, apex pointed, termen very obliquely rounded; fuscous irrorated with dark fuscous; a blackish dot at base of costa, and one almost at base above fold; a short very oblique whitish-ochreous streak from costa at preceded by a small blackish mark; a whitish-ochreous streak along fold from base to middle of wing, with a series of three small blackish marks, first on upper edge and the other two on lower; a very oblique whitish-ochreous streak from f of costa to § of disc, preceded by a series of three or four small blackish marks ; three small whitish-ochreous spots or dots on costa towards apex, and an undefined streak along termen, accompanied by a marginal series of small groups of blackish scales : cilia light grey, irrorated with whitish and somewhat sprinkled with blackish. Hindwings slaty-grey; cilia light grey. Natal, Umkomaas and Verulam, in January (Janse); two specimens. Gelechia fecunda n.sp. rj 9 mm. Head and thorax pale yellow-ochreous. Palpi whitish, second joint dark fuscous except apex, terminal joint shorter than second, with blackish subapical ring. Forewings elongate, rather narrow, costa gently arched, apex pointed, termen very obliquely rounded; yellow-ochreous; costa finely dotted with black from base to a triangular cloudy blackish spot at f ; a triangular spot of blackish-grey irroration on dorsum before tornus, slightly before costal spot ; a small black dot representing plical stigma ; some scattered blackish scales towards apex, and several undefined marginal dots of blackish irroration round apex and termen : cilia yellow-ochreous. Hindwings under 1, apex acutely produced ; rather dark grey ; cilia light grey. Natal, Umkomaas, in January (Janse); one specimen. Gelechia revoluta n.sp. cJ $ 10-11 mm. Head whitish sprinkled with dark fuscous. Palpi whitish, second joint with two bands of blackish irroration, terminal joint with apex and median band blackish. Thorax ochreous-whitish, slightly sprinkled with grey, shoulders and anterior margin more or less marked with dark fuscous. Abdomen grey, two basal segments whitish-ochreous. Forewings elongate, narrow, costa gently arched, apex tolerably pointed, termen extremely obliquely rounded; in $ with strong costal fold from base to near middle; whitish, irregularly and variably irrorated with grey; basal area suffused with pale ochreous; a blackish spot at base of costa; elongate blackish blotches on costa about ^ and dorsum before middle, more or less connected posteriorly by an oblique blackish blotch in disc; an elongate blackish blotch on costa 1-8 Annals of the Transvaal Museum before middle, beneath which is a round pale ochreous spot; stigmata black ringed with pale ochreous, plical slightly before first discal; a blackish spot on costa above second discal stigma, and a more or less developed blackish blotch on dorsum beneath and connected with it; apical area suffused with pale ochreous ; some cloudy black dots on posterior part of costa and termen : cilia dark grey sprinkled with whitish points. Hindwings i, apex considerably produced, pointed; grey; cilia grey. Zululand, Nkwaleni, in January (Janse); four specimens. Gelechia tetraleuca n.sp. $ 15 mm. Head and thorax white, patagia with a blackish stripe. Palpi white, base with some black scales. Abdomen whitish. Forewings elongate, narrow, costa gently arched, apex obtuse, termen rounded, very oblique; grey much suffused with black, especially through middle of disc, and towards dorsum near base; four white blotches, first elongate, extending along dorsum from J to f , widest anteriorly and reaching half across wing, thence irregularly attenuated, second oval, beneath costa slightly beyond middle, third roundish, on tornus, fourth semioval, on costa towards apex; a slender irregular white streak along termen: cilia white. Hindwings slightly over 1, apex obtuse, termen hardly sinuate; whitish-grey; cilia grey- whitish. Zululand, Eshowe, in January (Janse); one specimen. Phthorimaea geomicta n.sp. $ 12 mm. Head and thorax rather dark brown, shoulders and anterior part of dorsum suffused with blackish. Palpi brown irrorated with dark fuscous, terminal joint with median ring and tip whitish-mixed. Abdomen light grey. Forewings elongate-lanceolate; 6 separate; brown irrorated with fuscous, with some whitish scales; some darker fuscous suffusion towards base, and forming a patch in disc at J , a small blackish spot on costa at J , discal stigmata small, black, approximated: cilia whitish-grey, sprinkled with dark fuscous on basal half, with indistinct cloudy dark basal dots on costa. Hindwings pale bluish-grey; cilia whitish-grey. Natal, New Hanover, in October (Janse); one specimen. Phthorimaea pendens n.sp. $ 14 mm. Head white, crown with a few black specks. Palpi white, second joint tinged with ochreous and sprinkled with blackish, beneath with rather long rough hairscales, terminal joint loosely scaled, with two rings of blackish irroration. Thorax whitish, with a dot of black irroration in middle of anterior margin and three in a triangle posteriorly, patagia pale ochreous with two or three black specks. Abdomen ochreous-whitish. Forewings elongate, narrow, costa slightly arched, apex pointed, termen extremely obliquely rounded; reddish-ochreous, paler towards base, with a few scattered black scales; a blackish dot towards base above middle, one in middle beyond this, and one still further on beneath fold; an irregular oblique grey streak irrorated with black from costa at J to below middle, interrupted beneath costa, angulated and continued upwards to costa at § , and a similar streak from middle of costa crossing this and also below middle angulated upwards to touch it again just beyond second discal stigma; stigmata rather large, black, plical somewhat before first discal, these placed on margins of angle of first streak, second discal on posterior arm of first streak; some grey irroration along upper half of termen : cilia pale ochreous, sprinkled with black near base. Hindwings light slaty-grey; cilia whitish -ochreous. Annals of the Transvaal Museum *9 Zululand, Melmoth, in January (Jan%p); one specimen. The normal rough scaling of second joint of palpi is exaggerated into rather long hairscales in this species, and the loose scaling of terminal joint is also noticeable, but in all other particulars this species agrees well with Phthorimaea, and is pro- perly referred there. Polyhymno hostilis n.sp. $ 1 6 mm. Head white, centre of crown fuscous. Palpi white, second joint anteriorly dark fuscous. Thorax bronzy-brown, with five white stripes. Abdomen grey, anal tuft ochreous-whitish. Forewings elongate, narrow, costa slightly arched, apex strongly produced, acute, subfalcate, termen concave beneath apex, then obliquely rounded; bronzy-brown; a strong shining white median longitudinal streak from base to near termen, broadest beyond middle, whence it sends a line to termen above tornus, thence narrowed to a point ; a fine costal line almost from base to middle, thence very obliquely to just above apex of median streak; a narrow subdorsal white streak from base of dorsum to tornus; an oblique white line from f of costa, becoming silvery-metallic and angulated to form a short mark on termen beneath apex ; above posterior part of this an ochreous streak running into apical proj ection ; a wedgeshaped white spot before termen in middle, and some irregular white marking along termen : cilia white, on costa with two oblique dark fuscous marks at origin followed by a wedgeshaped dark fuscous spot, above apex with a basal dark fuscous mark followed by two oblique dark fuscous lines converging to apex, at apex with a grey anterior spot tipped with dark fuscous, beneath apex with a short black basal mark, below middle of termen with two dark fuscous subbasal dots, on tornal area light grey. Hindwings bluish- grey; cilia ochreous-whitish suffused with light grey towards base. Transvaal, Pretoria, in December (Janse); one specimen. Nearest pavacma, but with costal white line almost reaching base. Polyhymno intorta n.sp. 2 6 mm. Head ochreous-white. Palpi white, terminal joint with dark fuscous median band. Thorax dark fuscous with three white stripes. Fore- wings elongate, narrow, costa slightly arched, apex pointed, produced, termen sinuate beneath apex, then very obliquely rounded; blackish; two closely parallel whitish median lines from base, upper not reaching middle, lower continued along fold to tornus; a whitish dorsal line from base to tornus; a very oblique whitish streak from § of costa and a very oblique whitish line from ^ of dorsum, meeting at an acute angle in disc and produced to near termen, then shortly acute-angled back parallel to termen ; a fine double dark fuscous line suffused with fulvous from f of costa into apex, thence along termen to tornus, on costal portion margined on each side by fine whitish lines : cilia light grey, above apex with two wedgeshaped whitish marks separated and followed by black lines, at apex with a projecting black line, beneath apex with a short black basal mark and dark grey apical spot, beneath these with a whitish patch containing a slight brownish subbasal mark beneath. Hindwings violet-grey; cilia light grey. Natal, Umkomaas, in January (Janse); one specimen. The smallest species of the genus. Anacampsis elaeocoma n.sp. n-i2mm. Head whitish-ochreous. Palpi ochreous-whitish, second joint irrorated with dark fuscous on basal half, terminal joint with anterior 20 Annals of the Transvaal Museum edge irrorated with dark fuscous on apical half. Thorax fuscous, sometimes tinged with whitish-ochreous dorsally. Abdomen fuscous, two basal segments pale ochreous, anal tuft whitish. Forewings elongate-lanceolate, apex rather produced, acute; 6 separate; dark fuscous finely irrorated with whitish; stigmata blackish, discal approximated, plical obliquely before first discal : cilia pale greyish-ochreous, towards base irrorated with dark fuscous, with median line of black points. Hindwings i, apex very long-produced; light bluish-grey; cilia pale greyish-ochreous. Transvaal, Pretoria, in February and March (Janse); three specimens. Allied to cirrhocoma. Anacampsis exsulata n.sp. 12 mm. Head ochreous-white. Palpi white, second joint black except apex, terminal joint with a fine black line on each side except towards base. Thorax blackish, with large ochreous-whitish posterior spot. Abdomen grey. Forewings elongate, narrow, costa gently arched, apex pointed, termen sinuate, oblique; 6 separate; blackish; a moderately broad irregular-edged ochreous-whitish dorsal streak from base to tornus, its upper edge rather triangular-prominent at middle and f of length; a short oblique ochreous- white streak from costa at £ ; a triangular clear white spot occupying lower § of termen : cilia light grey irrorated with whitish, on costa dark grey with a basal line of white irroration becoming a subbasal white line on upper part of termen. Hindwings i, termen sinuate, apex pointed; grey; cilia light grey. Zululand, Eshowe, in January (Janse); one specimen. This very inter- esting insect is of South American type, quite unlike any Anacampsis from other regions ; the genus is very largely developed in South America. Acompsia eodryas n.sp. $ $ n-i2mm. Head whitish-ochreous. Palpi ochreous-whitish, second and terminal joints each with two bands of dark fuscous irroration. Thorax whitish-ochreous, patagia partially or almost wholly brown. Abdomen grey- whitish. Forewings elongate, narrow, costa slightly arched, apex pointed, termen extremely obliquely rounded; pale ochreous-yellowish, dorsal area variably tinged or mixed with brown; a narrow rosy-brown fascia from base of costa to dorsum before middle, marked with black on fold ; a similar fascia from £ of costa to dorsum before tornus, thence angulated upwards to costa before apex; the costal space enclosed by this forms a rosy-brown triangular blotch separated from it by a slender streak of ground-colour sometimes interrupted at its apex; stigmata black, on margin of angulated fascia, plical somewhat before first discal, second discal sometimes absent, a black mark also on dorsal angle of fascia ; costal edge of these markings more or less tinged with blackish ; an irregular line of black and rosy-brown scales along termen : cilia pale grey with rows of whitish points, on termen tinged with rosy and mixed with black towards base. Hindwings bluish-grey, lighter anteriorly; cilia light grey. Zululand, Nkwaleni, in January; Transvaal, Pretoria, in April (Janse) ; three specimens. The single Pretoria specimen sent has terminal joint of palpi only £ of second (instead of f), suffused with blackish, but as it is entirely similar otherwise, I regard the structure as probably an individual abnormality or deformity. The name Acompsia Hiibn. is here employed for the genus to which in my Handbook of British Lepidoptera the name Recurvaria Haw. was wrongly applied. Annals of the Transvaal Museum 21 Anarsia subfulvescens n.sp. $ io mm. Head and thorax whitish. Palpi dark fuscous, apical edge irrorated with white. Forewings elongate, rather narrow, costa gently arched, apex pointed, termen very obliquely rounded ; pale fuscous, finely and closely irrorated with white, with a few scattered black scales; an elongate black mark on middle of costa, and short fine black strigulae on costal edge near before and beyond it; some small groups of black specks on termen : cilia grey irrorated with whitish. Hindwings grey, thinly scaled anteriorly; cilia light grey. Wings beneath tinged with yellowish-fulvous towards base, especially hindwings, where it is prolonged on veins to middle. Natal, Umkomaas, in January (janse); one specimen. Anarsia vectaria n.sp. ? 8-13 mm. Head and thorax grey-whitish, shoulders narrowly black. Palpi white, second joint dark fuscous except apex, tuft yellowish, terminal joint with three dark fuscous rings. Abdomen in whitish, sides dark fuscous, anal tuft whitish-yellowish. Fore wings elongate, narrow, costa gently arched, apex obtuse, termen very obliquely rounded; pale grey finely irrorated with white; small black dots on costa at base and J, and one towards costa near base; a triangular blackish blotch extending on costa from ^ to f and nearly reaching fold, its apex subtruncate; a small black dot below fold near before apex of this; some indistinct blackish-grey dots round apical part of costa and termen : cilia grey sprinkled with whitish. Hindwings with costa in $ dilated on basal half; subhyaline-prismatic, posteriorly suffused with dark grey, veins dark grey; cilia light grey. .Forewings beneath with long whitish expansible hair pencil from base below cell, lying beneath expansion of hind- wings. Natal, Sarnia (Mrs Curry), Umkomaas (Janse), in November and January; two specimens. Anarsia spicata n.sp. cJ 13-14 mm. Head, palpi, and thorax light grey irrorated with whitish, patagia blackish, tuft of palpi suffused with dark fuscous towards apex. Abdomen light grey, sides blackish, anal tuft whitish, above brown. Fore- wings elongate, narrow, costa gently arched, apex tolerably pointed, termen very obliquely rounded; grey irrorated with whitish, with some scattered dark grey and blackish scales; a thick black submedian streak from base to middle, attenuated to a point, a median streak attenuated at both ends from before middle of disc to f , and an oblique dash from above apex of this to costa before apex : cilia light grey irrorated with whitish, round apex barred with dark grey basally. Hindwings grey, paler and thinly scaled anteriorly; cilia pale grey. Transvaal, Pretoria, in November and December (Janse) ; two specimens. Chelaria l oxo saris n.sp. cJ 17 mm. Head ochreous-whitish sprinkled with pale grey. Palpi ochreous- whitish, second joint with long broad subtriangular tuft, suffused with dark fuscous except along apical margin, terminal joint thickened with loose scales almost to apex, with two oblique subconfluent bands of blackish irroration above middle and a ring towards base. Thorax whitish irrorated with pale grey. Abdomen grey, anal tuft large, whitish-ochreous. Forewings elongate, rather narrow, costa moderately arched, apex obtuse, termen very obliquely 22 Annals of the Transvaal Museum rounded ; 6 separate ; fuscous finely and closely irrorated with white, with some scattered blackish scales; three or four blackish dots towards base; a small obliquely elongate dark fuscous spot on middle of costa, two small costal marks anterior to this and two posterior, all separated with whitish suffusion on costal edge; plical and second discal stigmata black; a series of cloudy dark fuscous marginal dots round posterior part of costa and termen : cilia fuscous or brownish irrorated with whitish, basal half indistinctly spotted with dark fuscous. Hindwings grey, darker posteriorly; cilia light fuscous. Forewings beneath with a long expansible fuscous hair-pencil from near base of dorsum extending obliquely upwards across disc. Natal, Umkomaas, in January (Janse); two specimens. Chelaria aust erodes n.sp. $ 1 6 mm. Head shining grey sprinkled with whitish. Palpi dark grey mixed with black and sprinkled with whitish, second joint with large broad tuft, with black median band and suffused with whitish basally, terminal joint with posterior scale-projection towards apex. Thorax blackish, patagia fuscous. Abdomen fuscous. Forewings elongate, narrow, costa gently arched, apex obtuse-pointed, termen extremely obliquely rounded; 6 separate; dark grey irregularly sprinkled with whitish, with scattered small blackish tufts and dashes; a longitudinal ochreous-brown mark near base in middle; seven or eight small oblique blackish spots along costa; an erect-oval spot of ground colour above tornus edged on upper half with blackish and on lower with whitish ; some blackish marking along termen : cilia dark grey sprinkled with whitish (imperfect). Hindwings dark grey, thinly scaled and whitish-tinged anteriorly, with hyaline streaks in and beneath cell; cilia light fuscous. Transvaal, Pretoria, in March (Janse); one specimen. Lachnostola n.g. Head smooth, sidetufts somewhat raised; ocelli small, far posterior; tongue developed. Antennae f , basal joint elongate, without pecten. Labial palpi long, recurved, second joint thickened with scales, forming a long projecting triangular apical tuft beneath, terminal joint as long as second, slender, acute. Maxillary palpi very short, filiform, appressed to tongue. Posterior tibiae clothed with hairs above. Forewings with i b furcate, 2 from f , 3 and 4 closely approximated from angle, 6 and 8 stalked, 7 absent, 1 1 from middle. Hindwings under 1, elongate-trapezoidal, apex produced, pointed, termen sinuate, cilia if; 3 and 4 connate, 5 nearly approximated, 6 and 7 stalked. Appears to be a development of Nothris. Lachnostola amphizeucta n.sp. $ 8-1 1 mm. Head and thorax whitish-yellowish, shoulders dark fuscous. Palpi yellow-whitish, second joint dark fuscous except towards apex. Abdomen grey. Forewings elongate, narrow, costa gently arched, apex pointed, termen extremely obliquely rounded ; pale ochreous-yellowish ; a narrow dark fuscous streak along costa from base to f ; a dark fuscous streak along dorsum from near base to f ; a deep ochreous spot surrounded with blackish suffusion resting on this before middle of wing; a triangular dark grey blotch extending on costa from middle to £ and reaching half across wing; discal stigmata blackish, second just below apex of costal blotch; more or less yellow-ochreous suffusion towards tornal area, with scattered blackish scales ; an apical spot of blackish Annals of the Transvaal Museum 23 suffusion : cilia pale ochreous-yellowish, with a dark grey spot above apical spot. Hindwings grey; cilia light grey. Natal, Umkomaas (Janse), New Hanover (Hardenberg) , in September, January, and March; three specimens. Dichomeris stromatias n.sp. $ 19 mm. Head and thorax crimson- whitish-grey. Palpi whitish-ochreous, second joint blackish except apex, with short rounded apical tuft beneath and scales expanded towards apex above, terminal joint somewhat longer than second. Abdomen grey, anal tuft pale ochreous. Forewings elongate, rather narrow, posteriorly slightly dilated, costa slightly arched, apex obtuse, termen straight, rather oblique; dull greyish-crimson; costal edge yellow- ferruginous from near base to near apex; stigmata indicated by whitish dots, second discal by two longitudinally placed, plical rather obliquely before first discal : cilia whitish-grey-crimson, more rosy- tinged towards base. Hind- wings grey, veins darker; cilia grey, base pale. Zululand, Nkwaleni, in January (Janse); one specimen. Trichotaphe conclusa n.sp. - of first List are doubtful species. A. rehmanniana Schinz J Erythrophloeum guineense Don (as far as material to hand). Mashishimale (Z.), March, T.M.H. 19,423. ' Cassia holosericea Fresen. Newington (Ptbg.), Rogers, 22,504; Moss and Rogers, 36. Pleiospora (Phaenohoffmannia) Bolusii Dumm. Houtbosch, Bolus, 10,995. P. cajanifolia Harv. Various localities. P. gracilior Dumm. McLea in Herb. Bolus, 5621. P. holosericea Schinz. Various localities. P. latebracteolata Dumm. Various localities. P. macrophylla Diimm. Macmac, Atherstone. P. paniculata Diimm. Houtbosch, Bolus, 11,034. P. obovata Schinz. Houtbosch, Rehmann, 6249; Waterval Boven, Feb., Rogers, 14,468. P. obovata Schinz var. brevepedunculata Diimm. Macmac, Atherstone. Lotononis Bainesii Baker. Various localities. L. calycina Bth. var. acuta Diimm. Lijdenburg, Wilms, 272, 273. L. calycina Bth. var. hirsutissima Diimm. Various localities. L. dichiloides Sond. Mavieriestad (E.), Dec., Mrs Pott, 5071. L. Gerrardi Diimm. var. transvaalensis. Kl. Olifant River, Schlechter, 3809. L. humilior Diimm. Lijdenburg, Wilms, 274. L. lanceolata Bth. Johannesburg, Rand, mi; Belfast, Dec., Miss Leendertz, 2866. L. laxa E. and Z. var. multiflora Diimm. Various localities. L. Marlothii Engl. Christiana, Nelson, 197. L. pulchra Diimm. Lijdenburg, Wilms, 280. L. sericoflora Diimm. Standerton, Rehmann, 6802; Standerton, Jan., Mrs Pott, 4101; Ermelo, Feb., Miss Leendertz, 3036. L. solitudinis Diimm. Carolina, Nov., Rademacher, T.M.H. 7271. L. transvaalensis Diimm. Various localities. L. Wilmsii Diimm. Between Middelburg and Crocodile River, Wilms, 277. L. Woodii Bolus. Standerton, Rehmann, 6794. Pearsonia aristata Diimm. Is Lotononis aristata Schinz of first List. P. Atherstonei Diimm. Various localities. P. filifolia Diimm. Is Lotononis filifolia Bolus of first List. P. marginata Diimm. Is Lotononis marginata of first List. P. multiflora Diimm. Is Lotononis multiflora Schinz of first List. P. multiflora Diimm, var. Stewartii Diimm. Swaziland, Miss Stewart. P. podalyriaefolia Diimm. Swaziland, Saltmarshe in Herb. Galpin, 989. P. propinqua Diimm. Lijdenburg, Wilms, 261; near Middelburg, Wilms, 261 c. P. Rogersii Diimm. Is Lotononis Rogersii Kensit of first List. P. sessilifolia Diimm. Is Lotononis sessilifolia Harv. of first List. P. swaziensis Diimm. Is Lotononis swaziensis Bolus of first List. Melolobium Wilmsii Harms. Between Drakensbergen and Pretoria, Wilms, 302 Crotalaria distans Bth. Silwane’s location, July, Dr Breyer, T.M.H. 18,652. C. grantiana Harv. Madsaba, March, Schlechter, 4585. C. lanceolata E. Mey. Barberton district, different collectors in T.M.H. C. maxillaris Klotzsch. Near Pretoria, March, Miss Leendertz, 761; Zeerust, Jan., Mrs Pott, 4369; Potchefstroom, March, Miss Leendertz, 3199. C. Monteiroi Taub. Selati, July, Dr Breyer, T.M.H. 18,653. C. natalitia Meisn. Barberton, Dec., Mrs Pott, 5337; Hlatikulu (Sw.), June, Miss Stewart, T.M.H. 8978. C. petiolaris Franch. var. australis Bak. f. Pietersburg, Bolus, 10,375 C. pilulicarpa Taub. var. Schinzii Bak. f. Pietersburg, Bolus, 10,915. C. Schinzii Bak. f. Sandriver, Schlechter, 4589. C. spartioides DC. Matebe Valley, Holub; Zeerust, Jan., Mrs Pott, 4258. C. spinosa Hochst. var. Schlechteri Bak. f. Magaliesberg, Nov., Schlechter, 3675; Beestkraal (R.), Dec., Jenkins, T.M.H. 6973. C. squarrosa Schinz var. Dinteri Bak. f. Transvaal, Schlechter, 4263. Argyrolobium longifolium Walp. Various localities. A. Muddii Diimm. Macmac, Mudd. 9—5 128 Annals of the Transvaal Museum Argyrolobium Wilmsii Harms. Lijdenburg, Wilms, 257. Bnchenroedera viminea Presl. Farm Nooit Gedacht (E.), Dec., Mrs Pott, 5070. Lotus nearest L. arabicus L. Messina, Dec., Rogers, 20,702. Indigofera rufescens E. Mey. Spitskop (E.), Dec.., Mrs Pott, 5083. I. stricta L. Pretoria, Jan., Miss Leendertz, 551. Psoralea Wilmsii Harms. Various localities. Teramnus labialis Spr. Lijdenburg, Wilms, T.M.H. 5890; Pretoria, Dec., Miss Leendertz, 460; Shilovane, March, Junod, T.M.H. 4930. Tephrosia macropoda E. Mey. Haenertsburg, Nov., Mrs Pott, 4636; Hlatikuku (Sw.), Miss Stewart, T.M.H. 9533; Nelspruit, Dec., Dr Breyer, T.M.H. 17,873. T. nubica Baker. Waterpoort (Z.), Sept., Rogers, 21,524. Sesbania punctata DC. Macoutsie, July, Dr Breyer, T.M.H. 18,664. S. tetraptera Hochst. Brakrivier, March, Schlechter, 4620. Lessertia perennans DC. var. sericea. Germiston, Rogers, 12,199 c. L. perennans DC. var. polystachya Harv. Is L. polystachya Harv. of first List. Astragalus burkeanus Bth. Koedoes Poort, May, Miss Leendertz, 627; Haenertsburg, Nov., Mrs Pott, 4519; The Downs (Ptgb.), Dec., Rogers and Moss. Desmodium dregeanum Bth. Shilovane, March, Junod, T.M.H. 5284. D. natalitium Sond. Barberton, Dec., Mrs Pott, 5334. Dalbergia melanoxylon Guill. and Perr. Komati Poort, Dec., Rogers, 22,171; Moss and Rogers, 580. Mucuna urens DC. Shilovane, Junod, 2393. Rhynchosia komatiensis Harms. Between Spitskop and Komati River, Wilms, 374. R. secunda E. and Z. Carolina, Dec., Rademacher, T.M.H. 10,491; Ermelo, Feb., Miss Leendertz, 3120; Rustenburg, Oct., Mrs Pott, 3431. Pueraria ficifolia L. Bolus. Is Rhynchosia ficifolia Bth. of first List. P. Rogersii L. Bolus. Selati Railway, Rogers, 11,806. GERANIACEAE. Geranium incanum L. Various localities. G. incanum L. var. Farm Goede Hoop, Dec., Mrs Pott, 4971. Pelargonium Bowkeri Harv. P. reniforme Curt. Wolmaransstad, Feb., Rogers, 22,704; Boksburg, Jan., Breyer, T.M.H. 15,519. P. zonale Willd. Waterval Onder, Rogers, 632. ERYTHROXYLACEAE. Erythroxylon caffrum Sond. Near Barberton, Oct., Thorncroft, 1045. ZYGOPHYLLACEAE. Tribulus Zeyheri Sond. On lands of Mapagoni (Z.), July, Dr Breyer, T.M.H. 16,044; Messina, March, Rogers, 19,373. CONN AR ACE AE. Cnetis natalensis Planch, and Sond. Haenertsburg, Nov., Mrs Pott, 4477. RUTACEAE. Fagara Davyi Verdoorn. Woodbush, Grenfell in Col. Herb. 1094; Pototato Bush (Ptbg.), Burtt-Davy, 1166; Eastwood in Col. Herb., 1298; Forbes’s Reef Bush (Sw.), Burtt-Davy, 2753. F. Thorncroftii Verdoorn. Barberton, Dec., Thorncroft in T.M.H. 9616. F. capensis Thunb. Is Xanthoxylon capense Harv. of first List. Thamnosma africanum Engl. The Downs (Ptbg.), Nov., Rogers, 22,000. Toddalia aculeata Lam. Modjadjes (Ptbg.), Dec., Rogers, 18,119. T. natalensis Sond. Waterpoort (Z.), Sept., Rogers, 21,542; Kaapsche Hoop, Aug., Rogers, T.M.H. 18,798. BURSERACEAE. Commiphora species. New species collected at Messina by Archdeacon Rogers. C. lugardae N.E.Br. Messina, March, Rogers, 20,762. MELIACEAE. Turraea obtusifolia Hochst. Rustenburg, Feb., van Dam, T.M.H. 16,517; Nelspruit, Annals of the Transvaal Museum 129 Dec., Dr Breyer, T.M.H. 17,692; Geelhoutkop (Wg.), Jan., Dr Breyer, T.M.H. 17,826; Kaapsche Hoop, March, Rogers, 20,828. T. nilotica Kotshy. Free State Mine (Z.), June, Dr Breyer, T.M.H. 17,571; The Downs, Nov., Rogers, T.M.H. 18,935 (in fruit). POLYGALACEAE. Polygala hispida Burch. Vlakfontein (E.), Rogers, 11,522. P. Rehmanni Chodat. P. rigens A. DC. Muraltia azorella Chod. Lijdenburg. M. conjuncta Chod. Swazieland, Galpin, 531. M. empetroides Chod. Houtbosch. EUPHORBIACEAE. Monadenium Lugardae N.E.Br. Great Letaba, near Birthday Road. June. Dr Breyer in T.M.H. Elaeophorbia acuta N.E.Br. Transvaal, Burtt-Davy. Euphorbia Bolusii N.E.Br. Near Middelburg, Bolus, 9767. E. clavigera N.E.Br. Near Bremersdorp (Sw.), Burtt-Davy, 3010. E. enormis N.E.Br. Pietersburg, Marloth, 5144 E. epicyparissias E. Mey. Is E. involucrata E. Mey of first List. E. Evansii Pax. Near Barberton, Evans; Potgietersrust, Burtt-Davy, 5657, E. glaucella Pax. Waterpoort (Z.), Rogers, 22,505. E. Gueinzii Boiss. Various localities. E. hirta L. Is E. pilulifera of first List. E. hypericifolia L. Shilovane, Junod, 644; Kaap Muiden, Thorncroft, 758; Hlatikulu (Sw.), July, Miss Stewart, T.M.H. 8868. E. inaequilatera Sond. Is E. sanguinea Hochst. of first List. E. ingens E. Mey. Near Barberton, Pole-Evans, 2919, 2931; Potgietersrust, Burtt- Davy, 2200, 5658. E.kraussianaBernh. var. /3. Houtbosch, Rehmann, 5913 ; Barberton, Dec., Thorncroft, 768; Mrs Pott, 5441; The Downs (Ptbg.), July, Rogers, 20,358. E. natalensis Bernh. Ermelo, Burtt-Davy, 5402; Tzaneen, Nov., Rogers, 12,580. E. neopolycnemoides Pax and Hoffm. Various localities. E. Tirucalli L. Moorddrift, Oct., Miss Leendertz, 2245; Potgietersrust, Madge, 8443; Marloth, 5146; Burtt-Davy, 1700; Mafutane, Bolus, 12,279; Komati Poort, June, Rogers. E. transvaalensis Schltr. Is E. Galpini Pax of first List. E. trichadenia Pax. Various localities. E. truncata N.E.Br. Various localities. Bridelia cathartica Bertol. Komati Poort, Kirk, 100; Rogers, T.M.H. 16,007; Barber- ton, Pole-Evans, 2945. B. mollis Hutch. Various localities. B. stipularis Blume in first List is an Indian species. Croton Gubouga S. Moore. Olifants River (Lg.), Pole-Evans; Pietersburg district (specimens in fruit in T.M.H.). C. subgratissimus Prain. Is C. gratissimus Burch, of first List. C sylvaticus Hochst. Modjadjes (Ptbg.), Rogers, 18,120. Fliiggea microcarpa Blume. Barberton, Thorncroft, Rogers, Mrs Pott; Potgietersrust Sept, and Jan., Miss Leendertz, 1244, 1940. Jatropha pseudoglandulifera Pax. Messina, Dec., Moss and Rogers, 78. Cluytia affinis Sond. Pietersburg and Lijdenburg districts. C. hirsuta E. Mey. Various localities. C. laxa Eckl. Lijdenburg, Wilms, 1318; Barberton, Galpin, 934. C. monticola S. Moore. Various localities. C. virgata Pax and Hoffm. Various localities. Acalypha angustata Sond. Many localities. A. caperonioides Baill. Many localities. A. caperonioides Baill. var. Galpini Prain. Barberton, Galpin, 1106. A. ciliata Forsk. Shilovane, Junod, 2188, 1028. A. depressinervia K. Schum Is A. Schinzii Pax of first List. 130 Annals of the Transvaal Museum Acalypha glabrata Thb. Crocodile River (Pr.), Feb., Miss Leendertz, 716; Barberton, Dec. Mrs Pott, 5507; Shilovane, Junod, 1100. A. glabrata Thb. var. latifolia Mull. Arg. Goedgedacht, March, Schlechter, 4602; Barberton, Jan., Miss Thorncroft, T.M.H. 4328; Krokodilpoort (Pr.), May, Dr Breyer, T.M.H. 13,852. A. glabrata Thb. var. pilosa Pax. Near Goedgedacht, Schlechter, 4602 partly. A. punctata Meisn. Various localities. A. punctata Meisn. var. longifolia Prain. Various localities. A. punctata Meisn. var. Rogersii Prain. Various localities. A. segetalis Mull. Arg. Various localities. A. senensis Klotz. Various localities. A. Wilmsii Pax. Various localities. Adenocline acuta (Thb.) Baill. Mavieriestad (E.), Nov., Mrs Pott, 4961; Barberton, Dec., Mrs Pott, 5491. Plukenetia africana Sond. Palalariver (Wg.), Jan.,. Dr Breyer, T.M.H. 18,071. Dalechampia Kirkii Prain. Komati Poort, Kirk, 60. Ctenomeria capensis Harv. Barberton, Dec., Mrs Pott, .5494. Tragia afhnis Mull. Arg. Vaal River, Zeyher, 1526. T. incisifolia Prain. Near Komati River, Bolus, 9779; Komati Poort, Schlechter, 11,781. T. natalensis Sond. Barberton, Dec., Mrs Pott, 5497. T. Rogersii Prain. Waterval Onder, Oct., Rogers, T.M.H. 2597. T. Sonderi Prain. Crocodile River, Burke; Magaliesberg, Zeyher and Burke; near ’Mbabane (Sw.), Bolus, 12,290. Phyllanthus Rogersi Hutch. Haenertsburg, Rogers, 19,023. ANACARDIACEAE. Heeria pulcherrima Oliv. The Downs (Ptbg.), Nov., Rogers, 22,005. Odina velutina Rich. Messina, Dec., Rogers, 22,119. Rhus Coriacea Engl. Is R. magalismontana Sond. of first List. R. gracillima Engl. Is R. filiformis Schinz of first List. R. villosa L.f. of first List is included in R. incana Mill. HIPPOCRATEACEAE. Hippocratea obtusifolia Roxb. Waterpoort (Z.), Sept., Rogers, 21,544. MELIANTHACEAE. Bersama transvaalensis Turrill. Barberton, Thorncroft, 817. RHAMNACEAE. Phylica paniculata Willd. Transvaal. VITACEAE. Cissus capensis Willd. Rietfontein (Ptbg.), Sept., Miss Leendertz, 833; Elandshoek, Sept., Rogers, T.M.H. 4848. C. fragilis E. Mey. Barberton, Dec., Mrs Pott, 5694. TILIACEAE. Corchorus hirsutus L. Geelhoutkop (Wg.), Jan., Dr Breyer, T.M.H. 17,821; Messina, Sept., Rogers, 19,356; Wyliespoort (Z.), Feb., Dr Breyer, T.M.H. 19,572. C. tridens L. Motepe, March, Schlechter, 4627; Rooiplaat (Pr.), Feb., Dr Breyer, T.M.H, 15,176 ; Barberton, Dec., Mrs Pott, 5629. Grewia rhytidophylla K. Schum. Messina, Moss and Rogers, 17,101. G. villosa Willd. Messina, Moss and Rogers, 77. MALVACEAE. Abutilon angulatum Mast. Shilovane, March, Junod, T.M.H. 5261; Ngelelle River (Z.), July, Dr Breyer, T.M.H. 16,024. Hibiscus aponeurus Spr. and Hutch. Joes Luck, near Barberton, Miss Thorncroft, Nov., T.M.H. 4989; Nelspruit, Dec., Dr Breyer, T.M.H. 17,961. H. diversifolius L. Selati Railway, Rogers, 2708, H. lunariifolius Wall. Various localities. H. surattensis L. Hlatikulu (Sw.), June, Miss Stewart, T.M.H. 13,953. H. ternatus Mast. Letsitele, Rogers, 2673 Annals of the Transvaal Museum 131 STERCULIACEAE. Melhania prostrata DC. Is M. rostrata DC. of first List. Dombeya rotundifolia Harv. Is D. densiflora of first List. Hermannia comosa Burch. Christiania, Nelson, 200. H. grisea Schinz. Near Brakriver, March, Schlechter, 4631. H. grosseserrata Schinz. Matye, March, Schlechter, 4628. H. macowanii (Szyszy.) Schinz. Marabastad, March, Schlechter, 4683. Stercuiia triphaca R.Br. Ngelelle River ( Z .), July, Dr Breyer, T.M.H. 16,009; Water- poort (Z.), Sept., Rogers, 21,535. ELATINACEAE; Bergia suffuticosa Fzl. Messina, Sept., Rogers, 19,303. CANELLACEAE. Warburgia Breyeri Pott. Near Macdutsie River, July, Dr Breyer, T.M.H. 17,573. FLACOURTIACEAE. Trimeria trinervis Harv. Mavieriestad (E.), Nov., Mrs Pott, 5115. Scolopia Mundtii (Arn.) Warb. Farm Nooit Gedacht (E.), Dec., Mrs Pott, 5097. S. Zeyheri Arn. Derde Poort (Pr.), July, Miss Leendertz, 190. TURNERACEAE. Wormskioldia glandulifera Klotzsch. Griffin Mine (Z.), Jan., Dr Breyer, T.M.H. 15,629; Messina, June, Rogers, 19,405. W. serrata Hochst. Maxaben, March, Schlechter, 4671; Messina, March, Rogers, 20,823. PASSIFLORACEAE. Tryphostemma arenophilum Pott. Warmbaths, Jan., Miss Leendertz, 2062. Adenia glauca Schinz. Magaliesberg, between Kameelpoort and Elandsriver, Reh- mann, 4799; Vygeboompoort (Wg.), Oct., van Dam, T.M.H. 13,715. A. multiflora Pott. Baviaanspoort (Pr.), Dr Fehrson, T.M.H. 13,786. A. Schlechteri Harms. A. stenophylla Harms. Lijdenburg, Dec., Wilms, 941. A. Wilmsii Harms. Lijdenburg, Nov., Wilms, 961; August, Wilms, 1897. THYMELAEACEAE. Lasiosiphon canoargentea C. H. Wright. Near Lijdenburg, Wilms, 1298; MacLea in Herb. Bolus, 3020. L. hoepfnerianus Vatke. Vereeniging, Leslie. L. similis C. H. Wright. Warmbaths, Sept., Miss Leendertz, 1314. L. Wilmsii C. H. Wright. Near the Vaal River, Wilms, 1299; near Crocodile River, Wilms, 1299 b. Arthrosolen polycephalus C. A. Mey. Sterkstroom River, Burke, 517; Bloemhof, July, Barrett-Hamelton, T.M.H. 6410. A. sericocephala Meisn. Various localities. A. variabilis C. H. Wright. Various localities. Gnidia fastigiata Rendle var. hirsuta Pearson. Johannesburg, Rand, 899. G. linoides Wikstr. Pretoria, Burtt-Davy, 685; Rustenburg, Miss Pegler, 977. G. ovalifolia Meisn. Macmac, Mudd. G. pinifolia L. Magaliesberg, Burke. LYTHRACEAE. Ammannia salsuginosa Guill. and Perr. Magalaquena River, April, Schlechter, 4771; Komati Poort, April, Rogers, T.M.H. 4795; Swaziland, Sept., Miss Stewart, T.M.H. 10,650; Nelspruit, Dec., Dr Breyer, T.M.H. 17,712. Nesaea sagittifolia (Sond.) var. glabrescens. Transvaal, Rehmann, 6796. . MYRTACEAE. Eugenia pusilla N.E.Br. Near Amsterdam (E.), Forbes. MELASTOMACEAE. Dissotis princeps Triana. New Agatha, June, Rogers, 18,890. 132 Annals of the Transvaal Museum HALORHAGIDACEAE. Serpicula repens L. Haenertsburg, Nov., Mrs Pott, 4759. UMBELLIFERAE. Alepidea basinuda Pott. Woodbush, Nov., Jenkins, T.M.H. 7443; Haenertsburg, Nov., Mrs Pott, 4656. A. comosa Diimm. Modderfontein, Conrath, 321. A. Jacobsziae Diimm. Lijdenburg, April, Jenkins, T.M.H. 10,331. A. Jenkinsii Pott. Middelburg, Nov., Jenkins, T.M.H. 9858. A. longeciliata Schinz. Between Middelburg and Crocodile River, Wilms, 567; Carolina, Nov., Rademacher, T.M.H. 8197. Bupleurum Mundtii Cham, and Schl. Zeerust, March, Jenkins, T.M.H. 11,662; Waterval Boven, June, Rogers; Crocodile River, Dec., Schlechter, 3917. OLEACEAE. Jasminum transvaalense Sp. Moore. Modjadjes (Ptbg.), Rogers, 18,108. ASCLEPIADACEAE. Raphionacme divaricata Harv. var. /3. Various localities. Tacazzea apiculata Oliv. var. benedicta. Komati Poort, Nov., Rogers, 12,633. Schizoglossum cordifolium E. Mey. Haenertsburg, Nov., Mrs Pott, 4650; Tzaneen, Nov., Rogers, 12,480. S. robustum Schltr. var. pubiflorum N.E.Br. Godwan River, Feb., Jenkins, T.M.H. i°.399; Spitskop (E.), Jan., Scheepers, T.M.H. 15,046; Pretoria, Jan., Miss Collins, T.M.H. 12,658. Pachycarpus macrochilus N.E.Br. Goede Hoop (E.), Dec., Mrs Pott, 4895. P. plicatus N.E.Br. Spitskop (E.), Dec., Mrs Pott, 5241. Sisyranthus imbertis Harv. Haenertsburg, Nov., Mrs Pott, 4649; Mavieriestad, Nov., Mrs Pott, 4901. Ceropegia multiflora Baker, var. latifolia N.E.Br. Pretoria, Jan., March, Mrs Pott, 4740, 4825. C. Thorncroftii N.E.Br. Barberton, Thorncroft; Kaap Valley, March, J. Thorncroft, T.M.H. 11,359. Tavaresia Barklyi N.E.Br. var. grandiflora. Louis Trichardt, Gettleffl, T.M.H. 11,459- Huernia carnosa Stent. Zilikats Nek, Magaliesberg, Pole-Evans, 11,020 in U.D.A. Herbarium. H. transvaalensis Stent. Crocodile River, Magaliesberg, Pole-Evans. H. zebrina N.E.Br. Near Komati Poort, van Dam, T.M.H. 15,635. Davalia polita N.E.Br. Zeerust, Jan., Mrs Pott, 4353. D. transvaalensis Schltr. var. parviflora L. Bolus. Near Naboomspruit, Galpin, 8467. Stapelia Gettleflii Pott. Louis Trichardt, Gettleffl, T.M.H. 9643. CONVOLVULACEAE. Dichondra repens Forst. Zeerust, Jan., Mrs Pott, 4170; Wolhuterskop, Nov., Miss Nunns, T.M.H. 18,772; Spelonken, Feb., Junod, 38. Ipomoea bathycolpos Hall. var. sinuato-dentata. Lijdenburg, Jan., Wilms, T.M.H. 5808; Geelhoutkop (Wg.), Jan., Dr Breyer, T.M.H. 17,799. I. magnusiana Schinz. Various localities. I. palmata Forsk. Waterval Boven, Nov., Rogers, 18,512; Blauwkop (Z.), Aug. Dr Breyer, T.M.H. 16,035. Jacquemontia capitata G. Don. Barberton, Dec., Mrs Pott, 5476; Lomato River, April, Jenkins, T.M.H. 9913. Merremia angustifolia Hook. f. Various localities. M. pinnata Hall. f. Nelspruit, Oct., Dr Breyer, T.M.H. 17,704. BORAGINEAE. Heliotropium zeylanicum Lam. Lekker Kraal, Jan., Krantz, T.M.H. 6421 ; Spelonken, Sept., Jenkins, T.M.H. 8157; Swaziland, July, Miss Stewart, T.M.H. 8814; Great Letaba, June, Dr Breyer, T.M.H. 17,576; Ngelelle (Z.), July, Dr Breyer, T.M.H. 16,038. Annals of the Transvaal Museum 133 VERBENACEAE. Lantana Camara L. Pretoria, Oct., Mrs Pott, 5710. Vitex mooiensis Pearson var. Rudolphi. Barberton, Dec., Mrs Pott, 5683; Nelspruit, Oct., Dr Breyer, T.M.H. 18,350. V. harveyana Pearson. Komati Poort, Nov., Rogers, T.M.H. 13,278. LABIATAE. Plectranthus Thorncroftii Sp. Moore. Barberton, J. Thorncroft, Herb. Rogers, 16,987. Thorncroftia longiflora N.E.Br. Joes Luck, near Barberton, April, Thorncroft, 795. Leucas glabrata R.Br. Various localities. Salvia africana L. Pilgrim’s Rest, July, Rogers. SOLANACEAE. Solanum bifurcum Hochst. Haenertsburg, Nov., Mrs Pott, 4498. S. aculeastrum Dun. Haenertsburg, Nov., Mrs Pott, 4653. S. nodiflorum Jacq. Barberton, Oct., Thorncroft, 1043. SCROPHULARIACEAE. Harveya crispula Conrath. Irene, Conrath, 966. Sphenandra viscosa Bth. of first List is a synonym of Sutera caerulea Hiern. BIGNONIACEAE. Markhamia acuminata K. Schum. Messina, March, Rogers, 20,767. PEDALIACEAE. Sesamum alatum Thonn. Swaziland, July, Miss Stewart, T.M.H. 8819; Nelspruit, Nov., Rogers, T.M.H. 3036; Messina, March, Rogers, 20,848. Sesamothamnus Lugardii N.E.Br. Messina, Dec., Rogers, 22,116. GESNERACEAE. Streptocarpus pusillus Harv. Volksrust, Jan., Jenkins, T.M.H. 9311; Belfast, Dec., Miss Leendertz, 2679; Mavieriestad (E.), Nov., Mrs Pott, 5005; Barberton, Dec., Mrs Pott, 5456. S. denticulata Turrill. Barberton, Feb., Thorncroft, T.M.H. 18,283. ACANTHACEAE. Thunbergia dregeana Nees. Nelspruit, Feb., Dr Breyer, T.M.H. 17,895. Hygrophila Rehmannii Schinz. Between Elandsriver and Klippan, Rehmann, 5056. Dyschoriste Rogersii Sp. Moore. Transvaal, Rogers, 12,876. Crabbea pedunculata N.E.Br. Waterval Boven, Nov., Rogers, 18,509. C. velutina Sp. Moore. Vygeboomspruit (Z.), June, Dr Breyer, T.M.H. 18,658. Blepharis Clarkei Schinz. Blauwberg, March, Schlechter, 4657; Warmbaths, Jan. Miss Leendertz, 2048; Majuba’s Kloof (Ptbg.), Nov., Mrs Pott, 4776. B. transvaalensis Schinz. Blauwberg, March, Schlechter, 4655; Rooiplaat (Pr ), March, Miss Leendertz, 784. Crossandra subacaulis C.B.C1. Tzaneen, Nov., Rogers, 12,583. Anisotes Rogersiae Sp. Moore. Messina, Rogers, 19,349. A. sessiliflorus C.B.C1. Between Ngelelle and Nuanetsi, July, Dr Breyer, T.M.H. 16,082; Messina, March, Rogers, 20,842. Dicliptera zeylanica Nees. Harmony Block (Ptg.), July, Dr Breyer, T.M.H. 17,568. Disperma transvaalense C.B.C1. Elandsriver, Rehmann, 4892. Ecbolium nearest E. amplexicaule. Komati Poort, Jan., Rogers, 22,488. Monechma Welwitschii C.B.C1. Messina, March, Rogers, 20,843. Rhinacanthus spec. Messina, Feb., Rogers, 22,599. RUBIACEAE. Oldenlandia macrophylla DC. Haenertsburg, Nov., Mrs Pott, 4818; Komati Poort, Nov., Rogers, T.M.H. 13,287; Nelspruit, Dec., Dr Breyer, T.M.H. 17,720. Gardenia resiniflua Hiern. Messina, Aug., March, Rogers, 19,489, 20,181. Tricalysia floribunda (Harv.) K. Schum. Komati Poort, Nov., Rogers, T.M.H. 13,284. T. bracteata Hiern. Waterpoort (Z.), Sept., Rogers. 134 Annals of the Transvaal Museum Pachystigma pygmaea Schltr. Is Vangueria pygmaea Schltr. of first List. Pavetta assimilis Sond. Various localities. Spermacoce natalensis Hochst. Various localities. DIPSACEAE. Cephalaria attenuata R. and S. var. /3. Standerton, Jan., Mrs Pott, 4052; Onderste Poort, Feb;, ex Theiler’s Lab.; Messina, March, Rogers, 20,862. Scabiosa transvaalensis Sp. Moore. Pilgrim’s Rest, Dec., Rogers, 14,361, 14,999; Spitskop, Wilms, 619. CUCURBITACEAE. Melothria acutifolia Cogn. Elandsriver, Rehmann, 4904, 4905. Cucumis africanus L. var. acutilobus Cogn. Transvaal, Rehmann, 5169, 6311. C. hirsutus Sond. var. dissectus Cogn. Transvaal, Rehmann, 6310. Trochomeria Hookeri Harv. var. quinquepartita Cogn. Transvaal, Rehmann, 6304. T. pectinata Cogn. var. subintegrifolia Cogn. Transvaal, Rehmann, 6309. Corrallocarpus sphaerocarpus Cogn. (3, scaberrimus Cogn. Boschveld, Rehmann, 4953, 5i70- C. sphaerocarpus Cogn. y, subhastatus Cogn. Klippan, Rehmann, 5160. Raphanocarpus spec. Griffin Mine (Z.), Jan., Dr Breyer, T.M.H. 19,725. C AMPANUL ACE AE . Wahlenbergia multiflora Conrath. Modderfontein, Conrath, 563. W. gracillima Sp. Moore. On the Selati River, between Komati Poort and Letaba River, Rogers, 2684. Cyphia Rogersii Sp. Moore. Modjadjes (Ptbg.), Rogers, 18,212. COMPOSITAE. Erlangea spec. Hlatikulu (Sw.), May, Miss Stewart, T.M.H. 12,872; The Downs (Ptbg.), Nov., Rogers, 21,932. Vernonia cineria Less. The Downs (Ptbg.), July, Rogers, 20,115. V. crataegifolia Hutch. Barberton, Galpin, 1350; Lijdenburg, April, Jenkins, T.M.H. 10,329. V. dregeana Sch. Bip. Nooit Gedacht (E.), Dec., Mrs Pott, 4880; Piet Retief, Oct., Miss Collins, T.M.H. 13,950. V. glabra Vatke. Various localities. V. mespilifolia Less. Barberton, Dec., Thorncroft, T.M.H. 9609; Hlatikulu (Sw.), March, Miss Stewart, T.M.H. 9632. V. pinifolia Less. Piet Retief, Dec., Jenkins, T.M.H. 10,950; Volksrust, Jan., Jenkins, T.M.H. 11,036. . V. primulina Hoffm. Harmony Block (Ptbg.), June, Dr Breyer, T.M.H. 17,566. V. Randii Sp. Moore. Various localities. V. Schlechteri O. Hoffm. Lijdenburg, Aug., Wilms. Dichrocephala latifolia DC. Lijdenburg, April, Wilms, T.M.H. 6446. Felicia elorigatus Thb. var. Messina, March, Rogers, 20,752. F. lutea N.E.Br. Various localities. Psiadia arabica J. and Sh. Messina, Rogers, 20,002. Nidorella depauperata Harv. Various localities. N. hirta DC. Lijdenburg, Jan., Wilms, T.M.H. 5804. Conyza pinnatifida Less. Groenkloof (R.), Dec., van Dam, T.M.H. 12,003. Blumea aurita DC. Ngelelle River (Z.), July, Dr Breyer, T.M.H. 16,089. Helichrysum epapposum Bolus. ’Mpone Berg, March, Schlechter, 4735. H. eriophorum Conrath. Irene, Conrath, 432. H. floccosum Klatt. Various localities. H. steneopterum DC. Various localities. H. umbraculigerum Less. Houtboschberg, March, Schlechter, 4733; Shilovane, Junod, 2340; Pilgrim’s Rest, April, Rogers, 18,554. H. undatum Less. Various localities. Anaglypha latifolia Sp. Moore. Pilgrim’s Rest, Dec., Rogers, 14,319. Pentatricha alata Sp. Moore. Pilgrim’s Rest, Rogers, 18,667. Monactinocephalus paniculatus Klatt. Houtbosch, Rehmann, 6068. Pegolettia senegalensis Cass. Great Letaba, June, Dr Breyer, T.M.H. 18,661. Annals of the Transvaal Museum 135 Calostephane divaricata Benth. Nuanetsi River, July, Dr Breyer, T.M.H. 16,053; Silwane’s location (Ptbg.), July, Dr Breyer, T.M.H. 17,567. Geigeria acaulis Benth. and Hook. Messina, Dec., Rogers, 22,265. G. aspera Harv. Various localities. Eclipta erecta L. Komati Poort, June, Rogers, T.M.H. 2345. Coreopsis Steppia Steetz. The Downs (Ptbg.), July, Rogers, 20,104. Thelesperma scabiosoides Less. Vereeniging, Nov., Mrs Pott, 3841. Lasiospermum radiatum Trev. Vereeniging, Nov., Mrs Pott, 3882. Matricaria multiflora Fzl. Sabie, Nov., Rogers, 18,604. M. nigellaefolia DC. Nuanetsi River (Z.), July, Dr Breyer, T.M.H. 16,059. M. pilifera Thell. Elandspruitbergen, Schlechter, 3846. Schistostephium saxicola Hutch. Johburg, Gilfillan in Herb. Galpin, 6218; Modder- fontein, Conrath, 402; Houtbosch, Rehmann, 6082. S. villosum Hutch. Swaziland, Hlatikulu, Miss Stewart, 77. Brachymeris athanasioides Hutch. Is Pentzia athanasioides S. Moore of first List. B. montana Hutch. Various localities. Lopholaena segmentata Sp. Moore. Lijdenburg, Jan., Wilms, T.M.H. 10,770; Barberton, Jan., J. Thorncroft, T M.H. 3962; Spitskop (E.), Dec., Mrs Pott, 5017. Cineraria lyrata DC. Standerton, Jan., Mrs Pott, 4076; Vereeniging, Nov., Mrs Pott, 3850. C. canescens Wendl. var. Standerton, Jan., Mrs Pott. Senecio Conrathii N.E.Br. Modderfontein, Conrath, 1202. S. isatideusDC. Lijdenburg, March, Wilms, T.M.H. 10,769; Zeerust, Jan., Mrs Pott, 4395; Spitskop (E.), Dec., Mrs Pott, 4876; Germiston, Nov., Rogers, 12,197. S. Johannesburgensis Sp. Moore var. dentatus Volksrust, Rogers, 19,007. S. latissimifolius Sp. Moore. Pilgrim’s Rest, Rogers, 14,946. S. macrophyllus Phillips. Houtbosch, Feb., Bolus, 10,993; Lijdenburg, Dec., Schlechter, 3956. S. Ommannei Sp. Moore. Johannesburg, Ommanney, in. S. oxyriaefolius DC. The Downs (Ptbg.), Nov., Rogers, 21,906. S. panduraefolius Harv. Hlatikulu (Sw.), March, Miss Stewart, T.M.H. 10,090; Lijdenburg, Nov., Wilms, T.M.H. 6468; Shilovane, April, Junod, T.M.H. 5282; Sabie, March, Rogers, 18,649 S. pinnulatus Thunb. Various localities. S. rhomboideus Harv. Spitskop (E.), Dec., Mrs Pott, 4875. S. subcoriaceus Schltr. Bethal, Dec., Mrs Pott, 3658. S. thyrsoideus DC. Graskop, Oct., Thorncroft, 979. S. urophyllus Conrath. Modderfontein, Conrath, 1202. S. vimineus DC. Potgietersrust, Oct., Miss Leendertz, 1235. Dimorphotheca caulescens Harv. Various localities. Tripteris auriculata Sp. Moore. The Downs (Ptbg.), Rogers, 20,243. Ursinia natalensis N.E.Br. Barberton, Jan., J. Thorncroft, T.M.H. 2830; Hlatikulu (Sw.), Nov., Miss Stewart, T.M.H. 9561. Venedium Bellidiastrum Sp. Moore. Standerton, Rogers, 18,758. Berkheyopsis Rehmannii Thell. Between Elandsriver and Klippan, Rehmann, 5078; Elandsriver, Rehmann, 4962; Makapansbergen, Streydpoort, Rehmann, 5453. Berkheya cousinioides Sp. Moore. Lijdenburg, Rogers, 14,546. B. polyacantha Sp. Moore. Standerton, Rogers, 18,459. Carduus pycnocephalus L. Volksrust, Jan., Jenkins, T.M.H. 9932. Gerbera speciosa Sp. Moore. Pilgrim’s Rest, Dec., Rogers, 14,322. G. natalensis Sch. Bip. of first List is a synonym of G. viridifolia Sch. Bip. CAMBRIDGE: PRINTED BY J. B. PEACE, M.A., AT THE UNIVERSITY PRESS 07. bV ^ ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM VOLUME VII Printed for the COMMITTEE OF THE TRANSVAAL MUSEUM BY THE UNIVERSITY PRESS, CAMBRIDGE, ENGLAND 1919-21 ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM VOLUME VII Printed for the COMMITTEE OF THE TRANSVAAL MUSEUM BY THE UNIVERSITY PRESS, CAMBRIDGE, ENGLAND 1919-21 PRINTED IN GREAT BRITAIN LIST OF CONTENTS VOLUME VII PAGE Dam, G. P. F. van Descriptions of New Species of Zonurus and Notes on the Species of Zonurus occurring in the Transvaal. (4 Plates) . . . .239 Description of a New Variety of a South African Lizard of the Family Geckonidae. (1 Plate) 244 Grobbelaar, C. S. The Scolopendridae of South Africa. (28 Text-figures) . . . 245 Hewitt, John A Short Survey of the Solifugae of South Africa. (8 Plates and 14 Text-figures) 1 Hoepen, E. C. N. van Contributions to the Knowledge of the Reptiles of the Karroo Formation: 5. A New Dinosaur from the Stormberg Beds. (2 Plates and 6 Text-figures) 77 6. Further Dinosaurian Material in the Transvaal Museum. (13 Plates and 27 Text-figures) . ...... 93 Descriptions of some Cretaceous Ammonites from Pondoland. (3 Plates) 142 Janse, A. J. T. On the South African Notodontidae, with Descriptions of apparently New Genera and Species. (14 Plates) ...... 149 Wagner, Percy A. Note on a Relic of the Phallus Cult among the M’Kahtla. (1 Plate) 262 INDEX TO VOLUME VII (New genera, subgenera, species, subspecies, and the main reference in a series of references in heavy-faced type; synonyms are in italics.) aemilianus, Holcodiscus, 146, 147 Aetonyx, 117, 118, 137 palustris, 117, 137 afra, Rhysida, 249 (Figs. 5, 6), 250 (Fig. 7), 251, 252 (Figs. 8, 9, 11, 12) africana, Henicops, 245, 260 africanus, Gryponyx, 87, 101, 102, 103, 118 africanus, Holcodiscus, 146; PI. XXVI, figs. 3-5 agramma, Phyllaliodes, 192; PI. I, figs. 22-24; PI. IV, figs. 18-19 Stauropus, 191 ajax, Pachysaurus, 102, 136 albicans, Pseudorethona, 168, 169; PI. I, fig. 9; PI. Ill, figs. 1-6 Rethona, 169 albicostata, Scalmicauda, 158, 159, 160; PI. I, fig. 1 ; PI. II, figs. 1-5 albida, Antheua, 176, 178 alcicornis, Solpuga, 14, 18, 32, 48; PI. VII, fig- 34 Alipes, 245, 246, 247 appendiculatus, 248 calcipes, 248 crotalus, 247 (Figs. 1-2) grandidieri, 248 multicostia, 248 ALIPINAE, 245 alnifolia, Trimeria, 163, 170 alstoni, Solpuga, 31, 47 amapondense, Gaudryceras, 143; PI. XXIV, figs. 4-5 Ammonites, 142 Ammosaurus, 91, 92 Amyops, 151, 152, 155, 205, 206 gigas, 206, 207; PI. V, figs. 11-14; PI. VII, figs. 18-19 ingens, 206, 207 Anaphe, 151, 153, 155, 232, 234 panda, 233 reticulata, 232, 233; PI. XI, figs. 9-16; PI. XII, fig. 1 anastomosis, Ichthyura, 161 anceps, Cormocephalus, 256 ANCHISAURIDAE, 80, 91, 92, 102 Anchisaurus, 91, 92 solus, 92 angolensis, Rana, 18 anodonta, Zana, 221, 222; PI. V, fig. 22 Antheua, 150, 151, 152 ftn., 153, 155 ftn., 157, 175, 196, 198, 201, 221, 222 Antheua, albida, 176, 178 aurifodinae, 176, 177, 179; PI. Ill, figs. . 27-30 basipuncta, 176, 178 bicolor, 175, 176, 180, 235; PI. I, fig. 20; PI. Ill, figs. 24-25 consanguinea, 177, 180 croceipuncta, 176, 177, 178; PI. I, fig. 19; PI. Ill, figs. 19-23 dimorpha, 155 ftn., 177, 180; PI. I, fig. 25; PL IV, figs. 14-17; PI. XIII, fig. 10 dimorpha var. brunnea, 181; PI. XIV, fig. 1 encausta, 176, 177, 179; PI. Ill, figs. 16-18 extenuata, 176 mixta, 176, 182; PI. XIII, fig. 9 peringueyi, 176, 182 simplex, 157, 175, 176, 177; PI. I, fig. 18; PI. Ill, fig. 26 tricolor, 175, 176, 177, 178, 179; PI. I, fig. 7; PI. Ill, figs. 10-15 varia, 179 Antheua, 176, 183, 222 Anticyra, 179, 180 antiquus, Thecodontosaurus, 91, 92 apiculatum, Combretum, 186 appendiculatus, Alipes, 248 Arctiomorpha, 232, 233 argenteomaculata, Scalmicauda, 16 1 argentescens, Cerura, 17 1, 17 1 ftn. aristata, Chaetacme, 202 Aristosaurus, 82, 84, 85, 86, 90, 92 erectus, 77, 78 (Fig. 1), 82 (Fig. 2), 84 (Fig. 3), 85 (Fig. 4), 86 (Fig. 5), 90 (Fig. 6), 92; Pis. IX-X Arthrorhabdus, 253, 259 formosus, 259 (Fig. 24) pygmaeus, 259 Asanada, 251, 253 socotrana, 253 (Figs. 13-14) Asteroscopus, 235 Atrasana, 203, 205, 235 atribasalis, Stauropus, 19 1 atrifrons, Scrancia, 21 1 atriguttata, Desmeocraera, 184, 186; PI. IV, fig. 8 Stauropus, 186 aurifodinae, Antheua, 175, 176, 177, 179, 198; PI. Ill, figs. 27-30 Rigema, 179 2 Annals of the Transvaal Museum austerus, Solpuga, 17 australis, Gluviopsis, 61 barberi, Chelypus, 70, 72 barbertonensis, Zonurus, 240; PL III basalis, Desmeocraera, 185, 186 basipuncta, Antheua, 176, 178 bechuanica, Solpuga, 14, 15, 39, 41, 51; PL IV, fig. 14 benga, Scalmicauda, 158 bernhardi, Daesia, 53 betschuanica, Daesia, 54, 56 bicolor, Antheua, 176, 180, 235; Pl. I, fig. 20; PL III, figs. 24-25 Chadisra , 180 bifasciata, Cerura, 171; Pl. XIII, fig. 8 bipars, Chadisra, 200 Blossia, 3, 5 (Text-fig. 1), 6, 9, 10, 12, 19, 22, 56, 59 clunigera, 59, 60 crepidulifera, 57, 60 echinata, 58, 59 falcicornis, 9 falcifera, 9, 57, 58 (Fig. 10 a), 60 falcifera, dolichognathus, 58 filicornis, 9, 58, 59, 60 fimbriata, 58, 61 karrooica, 57, 59 laminicornis, 57, 61 laticosta, 59, 60 litoralis, 57, 60 maraisi, 58, 59 namaquensis, 56, 61 obscura, 22, 59 setifera, 56, 57, 59 tricolor, 58, 60 unguicornis, 19, 56, 60; Pl. VII, figs. 39-40; Pl. VIII, figs. 44, 46 bouvieri, Hemiblossia, 61 Brachionycha, 235 punctulata, 235 Brachystegia, 160 ftn., 227 randii, 160 ftn., 227 bracteolata, Trema, 202 braunsi, Melanoblossia, 62 brevicornis, Cormocephalus, 257 brevipalpis, Solpuga, 17, 43 breyeri, Zonurus, 239, 241 ; Pis. I, II Breyeria, 15 1, 153, i57> 2I3 dasychiroides, 213, 214; Pl. V, fig. 16; Pl.VIII,figs.20-25;Pl.XIV, fig. 10 Broomiella , 7, 54 tineata , 54 Browni, Euskelesaurus, 100, 101 Massospondylus, 103, 118, 122 (Text- fig. 22), 123 (Text-figs. 23-24), 124 (Text-fig. 25), 125 (Text-fig. 26), 126 (Text-fig. 27) Thecodontosaurus, 117, 118, 138 brunnea, Antheua dimorpha, 181; PL XIV, fig. 1 brunnea, Eurystaura, 216; Pl. IX, figs. 5-9; Pl. XIV, fig. 11 bucephala, Phalera, 195, 196 buddhaicus, Holcodiscus, 146, 147 caffra, Solpuga, 43 calcaratus, Cormocephalus, 256 (Fig. 20), 257 calcipes, Alipes, 248 calliope, Desmeocraera, 184, 186; Pl. IV, figs. 4 and 5 Stauropus, 186 camelina, Lophopteryx, 166 Campyloctys, 151, 153, 158, 219 gladstonei, 219, 220; Pl. V, fig. 20; Pl. X, figs. 1-7; Pl. XIV, fig. 13 canescens, Desmeocraera, 185, 188; PL XIII, fig. 14 capensis, Euskelesaurus, 136, 137 Pachydactylus, 244 Scutigera, 245 Toreus, 63 Zonurus, 242 carinatus, Massospondylus, 91, 116, 117, 137, 138 catocaloides, Catochria, 235 Catochria, 235 catocaloides, 235 celeripes, Solpuga, 18, 31, 43, 47; Pl. VI, fig. 32 Ceroma, 10, 19, 23, 63, 64 focki, 64 inerme, 63, 65 leppanae, 64 pallidum, 63, 64 pictulum, 63 (Text-fig. 11), 64, 65; Pl. V, fig. 20 sclateri, 63, 65 Cerura, 150, 151, 155, 168, 169, 170, 171 ftn., 173, 174 argentescens, 17 1 bifasciata, 171; PL XIII, fig. 8 esmeralda, 17 1, 172 furcula, 170 marshalli, 171, 173 spiritalis, 170, 171, 172, 173; Pl. I, fig. 10; Pl. III, figs. 7-9; Pl. V, figs. 1-3 swierstrae, 170 ftn., 171, 173 cervina, Solpuga, 14, 15, 18, 32, 48 Chadisra , 180 Chadisra, 150, 151, 152, 153, 155, 155 ftn., 157, 180, 200, 213 bipars, 200 curvilinea, 200, 201; Pl. V. fig. 9; Pl. VI, figs. 26-28; Pl. VII, figs. 1-6 persimilis, 201, 202 rosinaria, 201, 202; PL VI, fig. 26 semiflava, 201, 202 uncifera, 201, 202 Chaetacme, 202 Index to Volume VII 3 Chaetacme aristata, 202 chelicornis, Solpuga, 8, 14, 16, 17, 20, 31, 38, 39, 41, 43, Si; PI- HI, fig. 11 Chelypus, 4, 6, 7, 18, 24, 64, 67, 68, 70, 72 barberi, 70, 72 hirsti, 70, 71 (Text-fig. 13), 72; PI. VIII, fig. 45 lennoxae, 70, 72 Chrysophyllum, 186 natalense, 186 viridifolium, 186 cinerea, Taeniopteryx, 212; PI. VIII, figs. 14-19; PI. XIV, fig. 9 clara, Epanaphe, 234 clarilla, Epanaphe, 234; PI. XII, figs. 9-14 Cleapa, 177 clunigera, Blossia, 59, 60 collinita, Solpuga, 14, 18, 33, 48 Colobopleurus, 253, 259 devylderi, 259, 260 (Figs. 25, 27) parcespinatus, 259, 260 (Figs. 26, 28) Combretum, 186, 195 apiculatum, 186 gueinzii, 195, 204 concolor, Hoplitis, 203, 204; PI. XIV, figs. 5, 6 congruata, Phyctimorpha, 208, 209; PI. XIV, fig. 7 consanguinea, Antheua, 177, 180 coquinae, Solpuga, 35, 37, 50 cordata, Eugenia, 190 cordylus, Zonurus, 242, 243 Cormocephalus, 245, 253, 255 anceps, 256 brevicornis, 257 calcaratus, 256 (Fig. 20), 257 dispar, 256 elegans, 255, 257 nitidus, 254 (Figs. 16-17), 255 (Figs. 18-19), 257 oligoporus, 257 pseudopunctatus, 257 setiger, 255, 256 COSSIDAE, 215 Cossus, 215 Crambometra, 151, 153, 158, 218, 220 derelicta, 218, 219, 220; PI. V, fig. 19; PI. IX, figs. 18-23; PI. XIV, fig. 12 crassimanus, Solpuga, 17 crassus, Hexisopus, 69 crepidulifera, Blossia, 57, 60 croceipuncta, Antheua, 176, 178; PI. I, fig. 19; PI. Ill, figs. 19-23 cro talus, Alipes, 247 (Figs. 1-2) CRYPTOPINAE, 245, 246 Cryptops, 245 cultrata, Solpuga, 38 cuneicornis, Solpuga, 52 Zeriassa, 52, 53 curvilinea, Chadisra, 200, 201; PI. V, fig. 9; PI. VI, figs. 27-28; PI. VII, figs. 1-6 Hyperaeschra , 201 cylindrodon, Thecodontosaurus, 91 cymosa, Dombeya, 202 Dacetum, 245 Daesia, 7, 9, 10, 12, 19, 20, 22, 53, 55, 68, 7i bernhardi, 53 betschuanica, 54, 56 hottentotta, 54, 56 kolbei, 54 leipoldti, 53, 54 lineata, 9 (Text-fig. 1 a), 53, 54, 55, 56; PI. VI, figs. 27, 31; PI- VIII, fig. 43 namaqua, 54, 55 pallida, 54 pearsoni, 56 rhodesiana, 55, 56 schreineri, 7, 54, 55 schultzei, 55, 56 subulata, 53, 55 DAESIINAE, 10, 20, 22, 66, 68 darlingi, Solpuga, 31, 44, 47 dasychira, Ramesa, 223 dasychiroides, Breyeria, 213, 214; PI. V, fig. 16; PI. VIII, figs. 20-25; PI. XIV, fig. 10 Hoplitis, 203, 204; PI. VII, figs. 7-12 Datana, 235 ministra, 235 ruficollis, 235 delalandi, Rana, 67 derbiana, Solpuga, 10, 11 (Text-fig. 1 b, B, C), 14, 16, 17, 18, 19, 33, 34, 35 (Text-fig. 7), 37, 42, 50; PI. II, fig. 4 derelicta, Crambometra, 218, 219; PI. V, fig. 19; PI. IX, figs. 18-23; PI- XIV, fig. 12 Desmeocraera, 151, 152, 156,183, 191, 194 atriguttata, 184, 186; PI. IV, fig. 8 basalis, 185, 186 calliope, 184, 186; PI. IV, figs. 4 and 5 canescens, 185, 188; PI. XIII, fig. 14 ianthina , 186 incana, 185, 189 interpellatrix, 183, 184, 185, 186; PI. I, fig. 21; PI. IV, figs. 1-3 octoginta, 188 pergrisea, 185, 19 1 platti, 185, 190; PI. XIII, fig. 16 steniptera, 185, 191; PI. IV, figs. 12 and 13 thalassina, 184, 187; PI. IV, fig. 9 tripuncta, 184, 189; PI. XIII, fig. 15 varia, 185, 187; PI. IV, fig. 7; PI. XIII, figs, n-13 vernalis, 184, 185, 186, 187; PI. IV, fig. 6 4 Annals of the Transvaal Museum Desmeocraera, 173, 174, 184 hierax, 173, 174, 184 Desmodium, 178 incanum, 178 devylderi, Colobopleurus, 259, 260 (Figs. 25, 27) dimorpha, Antheua, 155 ftn., 177, 180; PI. I, fig. 25; PI. IV, figs. 14-17; PI. XIII, fig. 10; PI. XIV, fig. 2 Dinara, 179 encausta, 179 discolor, Mimusops, 188, 19 1 dispar, Cormocephalus, 256 dissimilus, Pararethona hierax, 174 dolichognathus, Blossia falcifera, 58 (Text-fig. 10 b) Dombeya, 202 cymosa, 202 Dromicosaurus, 94, 103, 130, 138, 140 gracilis, 94, 95 (Text-fig. 2), 103, 105 (Text-figs. 8, 9), 106 (Text-figs. 10, 11), 107 (Text-fig. 12), 108 (Text- figs. 13, 14), no (Text-fig. 15), hi (Text-fig. 16), 112 (Text-fig. 17), 113 (Text-fig. 18), 1 14 (Text-fig. 19), 115 (Text-fig. 20), 1 16 (Text-fig. 21), 118, 130, 132, 138; PI. XIII, figs. 2-8; Pis. XIV-XVI echinata, Blossia, 58, 59 Ekebergia, 195 meyeri, 195 elegans, Cormocephalus, 255, 257 elegans, Leucophalera, 199; PI. V, fig. 8; PI. VI, figs. 22-25; PI- XIV, fig. 4 encausta, Antheua, 176, 177, 179; PI. Ill, figs. 16-18 Dinara, 179 Epanaphe, 151, 154, 155, 234 clara, 234 clarilla, 234; PI. XII, figs. 9-14 moloneyi, 234 ephippiata, Sirenopyga, 177 epigonum, Tetragonites, 144 erectus, Aristosaurus, 77, 78 (Fig. 1), 82 (Fig. 2), 84 (Fig. 3). 85 (Fig. 4), 86 (Fig. 5), 90 (Fig. 6), 92; Pis. IX-X erlenbergiensis, Plateosaurus, 102, 136 erythronota, Solpuga, 17, 33, 34, 49 erythronotoides, Solpuga, 17,33 (Text- fig- 6), 49 esmeralda, Cerura, 17 1, 172 Ethrnostigmus, 245, 247, 248 trigonopodus, 248 (Figs. 3, 4) Eucnemesaurus, 93, 99, 102, 116, 140 fortis, 93, 95 (Text-fig. 1), 96 (Text- figs- 3> 4), 97 (Text-fig. 5), 98 (Text- figs. 6, 7), 102, 1 18; Pis. XI; XII; XIII, fig. 1 Eucorybus, 245 Eugenia, 190 Eugenia cordata, 190 euprepiaeformis, Arctiomorpha, 233 Eurystaura, 15 1, 153, 158, 216 brunnea, 216; PI. IX, figs. 5-9; PI. XIV, fig. 11 Euskelesaurus, 100, 136, 137, 140 Browni, 100, 137 capensis, 136, 137 Eutimia, 183, 221, 222, 235 fagi, Stauropus, 194 Faku, Holcodiscus, 144; PI. XXV, figs. 3, 4; PI- XXVI, figs. I, 2 falcicornis, Blossia, 9 falcifera, Blossia, 9, 57, 58 (Text-fig. io«), 60 fasciata, Rana, 67 ferox, Solpuga, 13, 15, 18, 30, 31, 32 (Text-fig. 5), 47 filicornis, Blossia, 9, 58, 59, 60 fimbriata, Blossia, 58, 61 flavida, Pydna, 228 ftn., 229 focki, Ceroma, 64 fodiens, Hexisopus, 6, 69 forbesianum, Phylloceras, 142 formosus, Arthrorhabdus, 259 (Fig. 24) Pachydactylus capensis, 244 fortis, Eucnemesaurus, 93, 95 (Text- fig. 1), 96 (Text-figs. 3, 4), 97 (Text- fig- 5), 98 (Text-figs. 6, 7), 102, 1 18; Pis. XI, XII; XIII, fig. 1 furcifera, Solpuga, 14, 18, 25, 45 furcula, Cerura, 170 fusca, Solpuga, 12, 26, 44; PI. VI, fig. 26 fuscata, Polienus, 224, 225; PI. XIV, fig- 15 fuscigula, Rana, 18 fuscinota, Scalmicauda, 159 Galeodes, 69 GALEODIDAE, 20, 67, 68 Galona, 151, 152, 156, 157, 205 pyrrhotricha, 206 serena, 205, 206; PI. V, fig. 10; PI. VII, figs. 13-17 Gargetta, 21 1 Gaudryceras, 143 amapondense, 143 ; PI. XXIV, figs. 4, 5 GECKONIDAE, 244 GEOPHILIDAE, 261 GEOPHILUS, 261 giganteus, Zonurus, 240, 242 gigas, Amyops, 206, 207; PI. V, figs. 11- 14; PI. VII, figs. 18, 19 Hoplitis, 207 Melebaeas, 207 glabrifrons, Solpuga, 17 gladstonei, Campyloctys, 219, 220; PI. V, fig. 20; PI. X, figs. 1-7; PI. XIV, fig- 13 globiceps, Melanoblossia, 62 Index to Volume VII 5 globicornis, Solpuga, u, 13, 18, 27, 30, 46; PI. V, fig. 21 Gluviopsis, 19, 21 ftn., 23, 61 australis, 61 gracilis, Dromicosaurus, 94, 95 (Text- fig. 2), 98, 103, 105 (Text-figs. 8, 9), 106 (Text-figs. 10, ii), 107 (Text- fig. 12), 108 (Text-figs. 13, 14), no (Text-fig. 15), in (Text-fig. 16), 112 (Text-fig. 17), 1 13 (Text-fig. 18), 114 (Text-fig. 19), 1 15 (Text-fig. 20), 116 (Text-fig. 21), 1 18, 130, 132, 138; PI. XIII, figs. 2-8; Pis. XIV-XVI grandidieri, Alipes, 248 Gresslyosaurus, 101, 102, 118, 140 Plieningeri, 118 robustus, 1 01, 102 Grewia, 170 lasiocarpa, 170 griseitincta, Scalmicauda, 159, 160 griseiviridis, Stauropus, 19 1 Gryponyx, 87, 101, 102, 118, 140 africanus, 87, 101, 102, 103, 118 transvaalensis, 102, 140 gueinzii, Combretum, 195, 204 Gyposaurus, 91 hamata, Solpuga, 14, 15, 39, 51 HARPACTIRAE, 20 Harriesi, Massospondylus, 91, 102, 117, 136, 137 hastata, Solpuga, 12, 17, 42, 52 Hemiblossia, 5, 10, 19, 23, 61, 62 bouvieri, 61 idioceras, 62 kalaharica, 62 O’neili, 19, 61, 62; PI. Ill, fig. 5; PI. VIII, fig. 42 Hemicormocephalus, 251, 255 multispinus, 254 (Fig. 15), 255 Henicops, 245, 260 africana, 245, 260 Henosis, 232 heterogyna, Scalmicauda, 159; PI. I, fig. 2 Heterostoma, 245 HEXISOPODIDAE, 66 HEXISOPODINAE, 10, 20, 24, 68 Hexisopus, 5, 6, 10, 18, 24, 66, 71, 72 crassus, 69 fodiens, 6, 69 infuscatus, 69, 70 lanatus, 69, 71, 72; PI. VIII, fig. 41 nigrolunatus, 69, 70 reticulatus, 69, 70 hierax, Desmeocraera, 173, 174, 184, 191 Pararethona, 17 1 ftn., 174; PI. I, figs. 11-16 hirsti, Chelypus, 70 (Text-fig. 13), 72; PI. VIII, fig. 45 hirta, Protea, 172 Holcodiscus, 144, 146 Aemilianus, 146, 147 africanus, 146; PI. XXVI, figs. 3-5 buddhaicus, 146, 147 Faku, 144; PI. XXV, figs. 3-4; PI. XXVI, figs. 1, 2 Kandi, 146, 147 karapadensis, 146, 147 madrasinis, 146, 147 Hoplitis, 151, 152, 156, 203, 207, 235 concolor, 203, 204; PI. XIV, figs. 5, 6 dasychiroides, 203, 204; PI. VII, figs. 7-12 milhauseri, 203 phyllocampa, 203, 204 postica, 203, 204, 205 Hoplitis , 207 hostilis, Solpuga, 7, 8, 13, 14, 15, 16, 17, 18, 33, 36, 37, 38, 39, 4i, 42, 50; PI- IV, figs. 16, 17; PI. VI, fig. 29 hottentotta, Daesia, 54, 56 hyaenella, Zophodiopsis, 233 Hybocampa, 203 Hyperaeschra, 165, 201, 202 Hyperaeschra , 201 Hypophiala, 151, 152, 156, 192 melanogramma, 192, 193; PI. I, figs. 26, 27; PI. IV, figs. 20-24; PI. XIV, fig- 3 ianthina , Desmeocraera, 186 Ichthyura, 150, 15 1, 155, 161 anastomosis, 161 lentisignata, 162 roseotincta, 161, 162, 163; PI. I, fig. 6; PI. II, figs. 6-10; PI. XIII, fig. 4 violacearia, 162, 163; PI. XIII, figs. 5, 6 idioceras, Hemiblossia, 62 imitata, Phalera, 195, 196; PI. V, fig. 5; PI. VI, figs. 8-14 impedita, Stenostaura, 215; PI. V, fig. 17; PI. IX, figs. 1-4 impeditus, Cossus, 215 incana, Desmeocraera, 185, 189 incanum, Desmodium, 178 inerme, Ceroma, 63, 65 infuscatus, Hexisopus, 69, 70 ingens, Amyops, 206, 207 Inous, 235 interpellatrix, Desmeocraera, 183, 184, 185, 186; PI. I, fig. 21 ; PI. IV, figs. i-3 Stauropus , 185 jonesi, Zonurus, 243 junodi, Solpuga, 14, 15, 39, 48, 50 kafulica, Solpuga niassa, 73 (Text-fig. 14) kalaharica, Hemiblossia, 62 Kandi, Holcodiscus, 146, 147 6 Annals of the Transvaal Museum karapadensis, Holcodiscus, 146, 147 karrooica, Blossia, 57, 59 KARSCHIINAE, 20, 23, 68 keyserlingi, Solpuga, 18 kolbei, Daesia, 54 laminicornis, Blossia, 57, 61 lanatus, Hexisopus, 69, 70, 71, 72; PL VIII, fig. 41 lasiocarpa, Grewia, 170 lateralis, Solpuga, 17, 33, 34, 43, 49; PI. V, fig. 24 laticosta, Blossia, 59, 60 latimanus, Solpuga, 17 leipoldti, Daesia, 53, 54 lennoxae, Chelypus, 70, 72 lentisignata, Ichthyura, 162 leppanae, Ceroma, 64 lethalis, Solpuga, 12, 13, 14, 18, 24, 25 (Text-fig. 2 a), 26 Leucophalera, 15 1, 152, 157, 199 elegans, 199; PI. V, fig. 8; PI. VI, figs. 22-25; PI- XIV, fig. 4 Leviseuri, Pedeticosaurus, 77 lignitea, Phalera, 196, 197 lignosa, Zana, 221 LIMACODIDAE, 169 lineata, Broomiella, 54 Daesia, 9 (Text-fig. ia), 53, 54, 56; PI. VI, figs. 27, 28, 31; PI. VIII, fig. 43 Solpuga, 12, 17, 19, 42, 43, 52; PI. Ill, fig. 7; PI. VI, fig. 33 Lipophaga, 8, 18, 23, 65, 66 (Text-fig. 12) michaelseni, 65 schultzei, 65 trispinosa, 65 LITHOBIIDAE, 245, 260 litoralis, Blossia, 57, 60 Lophopteryx, 150, 151, 158, 165, 166 camelina, 166 saturata, 166 uniformis, 166, 167; PI. I, fig. 8; PI. II, figs. 18-23 lydenburgi, Phalera, 195, 196, 197; PI. V, fig. 6 LYMANTRIADAE, 150, 232 macer, Solpuga, 17 macrodonta, Ramesa, 222, 223; PI. V, fig. 23 ; PI. X, figs. 8-1 1 macrognathus, Solpuga chelicornis, 15, 41 (Text-fig. 9), 51 macropoda, Tephrosia, 18 1 madrasinus, Holcodiscus, 146, 147 magnus, Pachysaurus, 82 maraisi, Blossia, 58, 59 Solpuga, 17, 33, 49; PL V, fig. 23 marpissa, Eutimia , 183, 222, 235 Zana, 221, 222; PL V, fig. 21; Pl. IX, figs. 24-29 marshalli, Cerura, 17 1, 173 Solpuga, 14, 15, 16, 17, 36, 38, 39, 50; PL VI, fig. 30 Massospondylus, 91, 92, 102, 103, 116, 117, 118, 122, 123, 124, 125, 126, 136, 137, 138 Browni, 103, 118, 122 (Text-fig. 22), 123 (Text-figs. 23-24), 124 (Text- fig. 25), 125 (Text-fig. 26), 126 (Text- fig. 27); Pis. XVII-XXII carinatus, 91, 116, 117, 137, 138 Harriesi, 91, 102, 117, 136, 137 mediata, Ochrostigma, 195 Stauropus, 194, 195; Pl. V, fig. 4; PL VI, figs. 1-7 Melanoblossia, 9, 18, 22, 62 braunsi, 62 globiceps, 62 mclaoogramma, Hypophiala, 192, 193; PL I, figs. 26, 27; PL IV, figs. 20-24; PL XIV, fig. 3 Melebaeas, 206, 207 methueni, Solpuga, 42, 52 meyeri, Ekebergia, 195 michaelseni, Lipophaga, 65 milhauseri, Hoplitis, 203 Mimusops, 186, 188, 191 discolor, 188, 191 obovata, 186, 188 ministra, Datana, 235 minor, Thecodontosaurus, 92 mixta, Asitheua, 176, 182; Pl. XIII, fig. 9 modesta, Scrancia, 210 modestus, Polienus, 223, 224; PL X, figs. 12-17 moloneyi, Epanaphe, 234 monteiroi, Solpuga, 16, 19, 26, 27, 28, 45; Pl- VII, fig. 38 morsitans, Scolopendra, 257 (Figs. 21- 23), 258 multibracteata, Protea, 164, 172 multicostis, Alipes, 248 multispinus, Hemicormocephalus, 254 (Fig. 15), 255 MYRIAPODA, 245 namaqua, Daesia, 54, 55 namaquensis, Blossia, 56, 61 natalense, Chrysophyllum, 186 nera, Phylloceras, 142 Netria, 194 NEWPORTIIDAE, 245 niassa, Solpuga, 18, 73 nigrescens, Solpuga, 43 nigrobraccata, Solpuga, 43 nigrolunatus, Hexisopus, 69, 70 nigrosparsa, Polienus, 156, 224, 225; Pl. XIV, fig. 14 nitidiceps, Solpuga, 17 nitidus, Cormocephalus, 254 (Figs. 16, 17), 255 (Figs. 18, 19), 257 Index to Volume VII 7 niveiplaga, Scalmicauda, 161 NOCTUIDAE, 235 noctuiformis, Pectinophora, 164, 165; PL I, fig. 7; PL II, figs. 11-17; Pl. XIII, fig. 7 NOTODONTIDAE, 149, 15 1, 155, 232, 235 Notoxantha, 15 1, 153, 157, 217 sesamiodes, 217, 218; PL V, fig. 18; Pl. IX, figs. 10-17 nubeculosa, Brachionycha, 235 obovata, Mimusops, 186, 188 obscura, Blossia, 22, 59 Ochrostigma, 195 octoginta, Desmeocraera, 188 Stauropus, 19 1 oligoporus, Cormocephalus, 257 O’neili, Hemiblossia, 19, 61, 62; Pl. III, fig. 5; PL VIII, fig. 42 O’Neili, Prionocentrum, 226; PL X, figs. 18-22; Pl. XI, fig. 1; Pl. XIV, fig. 16 Scalmicauda, 158, 159, 160; PL I, figs. 4, 5; Pl. XIII, figs. 1-3 Opisthophthalmus, 17 orangicus, Solpuga coquinae, 37, 50 ornata, Rigema, 197, 198; PL V, fig. 7; Pl. VI, figs. 15-21 ornithorhyncha, Solpuga, 42, 52; PL V, fig. 19 Osica , 180, 235 verulama , 180, 235 OTOSTIGMINAE, 245, 246 Otostigmus, 246 Pachydactylus, 244 capensis formosus, 244 capensis tigrinus, 244; Pl. V Pachysaurus, 82, 102, 136 ajax, 102, 136 magnus, 82 pallida, Daesia, 54 pallidum, Ceroma, 63, 64 palustris, Aetonyx, 117, 137 panda, Anaphe, 233 Pararethona, 150, 15 1, 156, 17 1 ftn., 173, 184, 191 hierax, 171 ftn., 174; PL I, figs. 11-16 hierax dissimilus, 174 parcespinatus, Colobopleurus, 259, 260 (Figs. 26, 28) pearsoni, Daesia, 56 Pectinophora, 150, 151, 158, 164 noctuiformis, 164, 165; PL I, fig. 7; PL II, figs, n-17; Pl. XIII, fig. 7 Pedeticosaurus, 77 Leviseuri, 77 pergrisea, Desmeocraera, 185, 191 Stauropus, 19 1 peringueyi, Antheua, 176, 182 persimilis, Chadisra, 201, 202 petersi, Rhysida, 250, 251, 252 (Fig. 10) Phalera, 15 1, 152, 157, 195, 198, 199 bucephala, 195 imitata, 195, 196; Pl. V, fig. 5; PL VI, figs. 8-14 lignitea, 196, 197 lydenburgi, 195, 196, 197; PL V, fig. 6 Phalera, 198 Phycitimorpha, 1 51, 153, 156, 158, 208, 213 congruata, 208, 209; Pl. XIV, fig. 7 stigmatica, 208; Pl. VIII, figs. 1-5; Pl. XIV, fig. 8 Phyllaliodes, 15 1, 152, 156, 192 agramma, 192; PL I, figs. 22-24; Pl. IV, figs. 18, 19 phyllocampa, Hoplitis, 203, 204 Phylloceras, 142, 143 Forbesianum, 142 Nera, 142 Rogersi, 143 umzambiense, 142; PL XXIV, figs. 1- 3 Velledae, 143 picta, Solpuga, 12, 43 pictulum, Ceroma, 63 (Text-fig. 11), 65; Pl. V, fig. 20 PLATEOSAURIDAE, 80, 91, 92, 102, 120, 135, 140 Plateosaurus, 89, 101, 102, 116, 118, 135, 136, 140 erlenbergiensis, 102, 136 poligniensis, 102 Quenstedti, 118, 135, 136, 140 Reinigeri, 89, 102, 135 stormbergensis, 102, 116 platti, Desmeocraera, 185, 190; Pl. XIII, fig. 16 Plieningeri, Gresslyosaurus, 118 plumitarsus, Polelassothys, 229, 230 Pl. XII, figs. 2-8; Pl. XIV, fig. 17 Polelassothys, 15 1, 153, 155, 229 plumitarsus, 229, 230; PL XII, figs. 2- 8; Pl. XIV, fig. 17 Polienus, 15 1, 153, 156, 223 fuscata, 224, 225; Pl. XIV, fig. 15 modestus, 223, 224; Pl. X, figs. 12-17 nigrosparsa, 156, 224, 225; Pl. XIV, . fi?- 14 poligniensis, Plateosaurus, 102 poliostrota, Somera, 235 postica, Atrasana, 204, 205, 235 Hoplitis, 203, 204, 235 Prionocentrum, 15 1, 153, 157, 226 O’neili, 226; Pl. X, figs. 18-22; PL XI, fig. 1; PL XIV, fig. 16 Pro tea, 164, 172 hirta, 172 multibracteata, 164, 172 Pseudoblossia, 65 8 Annals of the Transvaal Museum pseudopunctatus, Cormocephalus, 257 Pseudorethona, 150, 151, 156, 168 albicans, 168, 169, 173, 174; PL I, fig. 9; PL III, figs. 1-6 pubescens, Solpuga chelicornis, 40, 41, 42, 51; PI- II, fig. 2 pugalitor, Solpuga, 28 punctulata, Brachionycha, 235 purcelli, Zeriassa, 52, 53; Pl. VII, fig. 36 Pydna, 151, 153, 156, 157, 228 flavida, 228 ftn., 229 rubrifascia, 153, 157, 228 ftn., 229; Pl. XI, fig. 3 rubritincta, 156, 228, 229; Pl. XI, figs. 2, 4-8 testacea, 228 pygmaeus, Arthrorhabdus, 259 pyrrhotricha, Galona, 206 Quenstedti, Plateosaurus, 118, 135, 136, 140 Ramesa, 15 1, 153, 157, 222 dasychira, 223 macrodonta, 222, 223; PL V, fig. 23; Pl. X, figs. 8-1 1 tosta, 222 Rana, 18, 67 angolensis, 18 delalandii, 67 fasciata, 67 fuscigula, 18 randii, Brachystegia, 160 ftn., 227 rectus, Solpuga lethalis, 14, 18, 24, 25 (Text-fig. 2b), 44; Pl. VII, fig. 37 Reinigeri, Plateosaurus, 89, 102, 135 Rethona, 169 albicans, 169 strigosa, 169 reticulata, Anaphe, 232, 233; PL XI, figs. 9-16; Pl. XII, fig. 1 reticulatus, Hexisopus, 69, 70 Rhagodes, 68, 69 ftn. RHAGODINAE, 68 rhodesiana, Daesia, 55, 56 Solpuga, 73 Rhus, 180 villosa, 180 Rhysida, 245, 247, 248, 252 afra, 249 (Figs. 5, 6), 250 (Fig. 7), 251, 252 (Figs. 8, 9, 11, 12) petersi, 250, 251, 252 (Fig. 10) Rigema , 179 Rigema, 151, 152, 157, 179, 195, 196, 197 ornata, 197, 198; Pl. V, fig. 7 vittata, 197 woerdeni, 198 robertsi, Zonurus, 241; Pl. IV robustus, Gresslyosaurus, 101, 102 Rogersi, Phylloceras, 143 roseotincta, Ichthyura, 161, 162, 163; Pl. I, fig. 6; Pl. II, figs. 6-10; Pl. XIII, fig. 4 rosinaria, Chadisra, 201, 202; Pl. VI, fig. 26 rubrifascia, Pydna, 153, 157, 228 ftn., 229; Pl- XI, fig. 3 rubritincta, Pydna, 156, 228, 229; Pl. XI, figs. 2, 4-8 rufescens, Solpuga chelicornis, 40, 41, si; pi- 11, fig- 3 ruficollis, Datana, 235 rugosa, Scutigera, 245 sagittaria, Solpuga, 18, 31, 47 saturata, Lophopteryx, 166 Sauropoda, 97 Scalmicauda, 150, 151, 155, 158 albicostata, 158, 159, 160; Pl. I, fig. 1; Pl. II, figs. 1-5 argenteomaculata, 161 benga, 158 fuscinota, 159 griseitincta, 159, 160 heterogyna, 159; PL I, fig. 2 niveiplaga, 161 O’neili, 158, 159, 160; Pl. I, figs. 4, 5; PL XIII, figs. 1-3 schlechteri, Solpuga, 18, 27, 45; Pl. IV, fig- 15 schonlandi, Solpuga, 13, 16, 18, 29 (Text- fig- 4), 46, 73 l schreineri, Daesia, 7, 54, 55 schultzei, Daesia, 55, 56 Lipophaga, 65 Solpuga, 13, 43 schweinfurthi, Solpuga, 18 sclateri, Ceroma, 63, 65 SCOLOCRYPTOPIDAE, 245 Scolopendra, 245, 253, 257, 258 morsitans, 257 (Figs. 21-23), 258 SCOLOPENDRIDAE, 245, 246, 260 SCOLOPENDRINAE, 245, 246, 251, 258 scopulata, Solpuga, 43 Scrancia, 15 1, 153, 156, 208, 210 atrifrons, 21 1 modesta, 210 stictica, 210, 21 1 ; PL V, fig. 15; Pl. VIII, figs. 6-13 Scutigera, 245 capensis, 245 rugosa, 245 semiflava, Chadisra, 201, 202 serena, Galona, 205, 206; PL V, fig. 10; PL VII, figs. 13-17 sericea, Solpuga, 13, 16, 18, 31, 43, 44, 47; Pl- HI, fig. 6 serraticornis, Solpuga, 14, 18, 27, 46 sesamiodes, Notoxantha, 217, 218; Pl. V, fig. 18; Pl. IX, figs. 10-17 Index to Volume VII 9 setifera, Blossia, 56, 57, 59 setiger, Cormocephalus, 255, 256 simplex, Antheua, 157, 175, 176, 177; PI. I, fig. 18; PI. Ill, fig. 26 Sirenopyga, 175, 177 ephippiata , 177 skirtopodus, Thecodontosaurus, 92, 117, 138 socotrana, Asanada, 253 (Figs. 13, 14) SOLIFUGAE, 3, 4, 12, 19, 20, 21, 67, 69 ftn. Solpuga, 7, 8, 9, 10, 11, 12, 13 18, 19, 21, 24, 43, 67, 71, 73 alcicornis, 14, 18, 32, 48; PI. VII, fig. 34 alstoni, 31, 47 austerus, 17 bechuanica, 14, 15, 39, 41, 51; PI. IV, fig. 14 brevipalpis, 17, 43 caffra, 43 celeripes, 18, 31, 43, 47; PL VI, fig. 32 cervina, 14, 15, 18, 32, 48 chelicornis, 8, 14, 16, 17, 20, 31, 38, 39, 41, 43, 51; PI- HI, fig. 11 chelicornis macrognathus, 15, 41 (Text- fig. 9), 5i chelicornis pubescens, 40, 41, 42, 51; PI. II, fig. 2 chelicornis rufescens, 40, 41, 51; PI. II, fig- 3 collinita, 14, 18, 33, 48 coquinae, 35, 37, 50 coquinae orangicus, 37 coquinae typicus, 50 crassimanus, 17 cultrata, 38 cuneicornis, 52 darlingi, 31, 44, 47 derbiana, 10, 11 (Text-fig. 1 b, B, C ), 14, 16, 17, 18, 19, 33, 34, 35 (Text- fig. 7), 37, 42, 50; PL II, fig. 4 erythronota, 17, 33, 34, 49 erythronotoides, 17, 33 (Text-fig 6), 49 ferox, 13, 15, 18, 30, 31, 32 (Text-fig. 5), 47 furcifera, 14, 18, 25, 45 fusca, 12, 26, 44; Pl. VI, fig. 26 glabrifrons, 17 globicornis, 11 (Text-fig. 1 b, A), 13, 18, 27, 30, 46; Pl. V, fig. 21 hamata, 14, 15, 39, 51 hastata, 12, 17, 42, 52 hostilis, 7, 8, 10, 13, 14, 15, 16, 17, 18, 33, 36, 37, 38, 39, 4i, 42, So; Pl. IV, figs. 16, 17; PL VI, fig. 29 junodi, 14, 15, 39, 48, 50 keyserlingi, 18 lateralis, 17, 33, 34, 43, 49; Pl. V, fig. 34 latimanus, 17 lethalis, 12, 13. 14, 18, 24, 25 (Text- fig. 2 a), 26 Solpuga lethalis rectus, 14, 18, 24, 25 (Text-fig. 2 b), 44; Pl. VII, fig. 37 lethalis typicus, 14, 24, 44 lineata, 12, 17, 19, 42, 43, 52; Pl. III, fig. 7; Pl. VI, fig. 33 macer, 17 maraisi, 17, 33, 49; Pl. V, fig. 23 marshalli, 14, 15, 16, 17, 36, 38, 39, 50; Pl. VI, fig. 30 methueni, 42, 52 monteiroi, 16, 19, 26, 27, 28, 45; PL VII, fig. 38 niassa, 18, 73 niassa kafulica, 73 (Text-fig. 14) nigrescens, 43 nigrobraccata, 43 nitidiceps, 17 ornithorhynchus, 42, 52; PL V, fig. 19 picta, 12, 43 pugalitor, 28 rhodesiana, 73 sagittaria, 18, 31, 47 schlechteri, 18, 27, 45; PL IV, fig. 15 schonlandi, 13, 16, 18, 29 (Text-fig. 4), 46, 73 schultzei, 12, 43 schweinfurthi, 18 scopulata, 43 sericea, 13, 16, 18,31,43,44, 47; Pl. Ill, fig. 6 serraticornis, 14, 18, 27, 46 serraticornis umtalica, 28, 46 spectralis, 17, 33, 49 spiralicornis, 14, 18, 28, 29, 45, 46; PL V, fig. 18 strepsiceros, 14, 18, 28, 29 (Text-fig. 3), 46; PL IV, fig. 13 striata, 18, 43 suffusca, 10, 32, 47; Pl. V, fig. 22 tookei, 14, 35, 36 (Text-fig. 8), 50 toppini, 26, 43, 44 tubicen, 39, 50; PL VII, fig. 35 Venator, 8, 12, 13, 14, 16, 17, 18, 19, 22, 25, 26, 27, 30, 39, 45; Pl. I, fig. 1; Pl. IV, fig. 12 venosa, 25, 44 villosa, 16, 17, 39, 41, 42, 51 vincta, 17, 33, 49 zebrina, 18, 43 SOLPUGIDAE, 67, 68 SOLPUGINAE, 20, 21 solus, Anchisaurus, 92 Somera, 235 poliostrota, 235 spectralis, Solpuga, 17, 33, 49 sphinx (cassinea), Brachionycha, 235 spiralicornis, Solpuga, 14, 18, 28, 29, 45, 46; Pl. V, fig. 18 spiritalis, Cerura, 170, 17 1, 17 1 ftn., 172, 173; PL I, fig. 10; Pl. III, figs. 7-9; PL V, figs. 1-3 10 Annals of the Transvaal Museum spurcata, Antheua , 176, 183, 222 Zana, 235 Stasimopus, 10, 12, 67 Stauropus, 151, 152, 155, 184, 191, 194 agramma, 19 1 atribasalis, 19 1 atriguttata, 186 fagi, 194 griseiviridis, 19 1 interpellatrix, 185 mediata, 184, 194, 195; PL V, fig. 4; PL VI, figs. 1-7 octoginta, 19 1 pergrisea, 19 1 steniptera , 19 1 thalassina, 187 steniptera, Desmeocraera, 185, 191; Pl. IV, figs. 12, 13 Stauropus , 191 Stenostaura, 151, 153, 155, 212, 215, 216 impedita, 215; Pl. V, fig. 17; Pl. IX, figs. 1-4 stictica, Scrancia, 210, 211; Pl. V, fig. 15; PL VIII, figs. 6-13 stigmatica, Phycitimorpha, 208; Pl. VIII, figs. 1-5; Pl. XIV, fig. 8 stormbergensis, Plateosaurus, 102, 116 strepsiceros, Solpuga, 14, 18, 28, 29 (Text-fig. 3), 46; Pl. IV, fig. 13 striata, Solpuga, 18, 43 strigosa, Rethona, 169 STRIPHNOPTERY GIDAE, 152, 232, 233 subulata, Daesia, 53, 55 suevicus, Teratosaurus, 118 suffusca, Solpuga, 10, 32, 47; Pl. V, fig. 22 swierstrae, Cerura, 170 ftn., 171, 173 Taeniopteryx, 15 1, 153, 158, 212 cinerea, 212; Pl. VIII, figs. 14-19; Pl. XIV, fig. 9 Tephrosia, 18 1 macropoda, 18 1 Teratosaurus, 118, 140 suevicus, 1 18 teres, Tetragonites, 144; Pl. XXV, figs. 1, 2 testacea, Pydna, 228 Tetragonites, 144 epigonum, 144 teres, 144; Pl. XXV, figs. 1, 2 Timotheanum, 144 Thacona , 21 1 thalassina, Desmeocraera, 184, 187; PL IV, fig. 9 Stauropus, 187 Thecodontosaurus, 91, 92, 117 antiquus, 91, 92 Browni, 117, 118, 138 cylindrodon, 91 minor, 92 Thecodontosaurus skirtopodus, 92, 117, 138 Theropoda, 80, 97, 101, 103, 140 tigrinus, Pachydactylus capensis, 244; Pl. V Timotheanum, Tetragonites, 144 tookei, Solpuga, 14, 35, 36 (Text-fig. 8), 50 toppini, Solpuga, 26, 43, 44 Toreus, 18, 23, 63 capensis, 63 tosta, Ramesa, 222 Trachycormocephalus, 253, 255 transvaalensis, Gryponyx, 102, 140 Trema, 202 bracteolata, 202 Trematoptychus, 245 tricolor, Antheua, 175, 176, 177, 178, 179, 180; Pl. I, fig. 17; Pl. Ill, figs. 10-15 Blossia, 58, 60 trigonopodus, Ethmostigmus, 248 (Figs. 3, 4) Trimeria, 163, 170 alnifolia, 163, 170 tripuncta, Desmeocraera, 184, 189; PL XIII, fig. 15 trispinosa, Lipophaga, 65 tropidogaster, Zonurus, 243 tubicen, Solpuga, 39, 50; Pl. VII, fig. 35 typicus, Solpuga coquinae, 59 Solpuga lethalis, 14, 24, 44 umtalica, Solpuga serraticornis, 28, 46 umzambiense, Phylloceras, 142; Pl. XXIV, figs. 1-3 uncifera, Chadisra, ,201, 202 unguicornis, Blossia, 19, 56, 60; Pl. VII, figs. 39, 40; PL VIII, figs. 44, 46 uniformis, Lophopteryx, 166, 167; Pl. I, fig. 8; Pl. II, figs. 18-23 Uroplectes, 27 vittatus, 27 varia , Antheua, 179 varia, Desmeocraera, 185, 187; Pl. IV, fig. 7; PL XIII, figs. 11-13 Velledae, Phylloceras, 143 Venator, Solpuga, 8, 12, 13, 14, 16, 17, 18, 19, 22, 25, 26, 27, 30, 39, 45, 74; PL I, fig. 1; Pl. IV, fig. 12 venosa, Solpuga, 25, 44 vernalis, Desmeocraera, 184, 185, 186, 187; Pl. IV, fig. 6 verulama, Osica, 180, 235 villosa, Rhus, 180 Solpuga, 16, 17, 39, 41, 42, 51 vincta, Solpuga, 17, 33, 49 violacearia, Ichthyura, 162, 163; Pl. XIII, figs. 5, 6 viridifolium, Chrysophyllum, 186 vittata, Rigema, 197 Index to Volume VII 1 1 vittatus, Uroplectes, 27 vittifer, Zonurus, 242, 243 warreni, Zonurus, 241 woerdeni, Phalera , 198 Rigema, 198 Zana, 151, 157, 177, 196, 213, 221, 232, 235 anodonta, 221, 222; Pl. V, fig. 22 lignosa, 221 marpissa, 221, 222; PI. V, fig. 21; PI. IX, figs. 24-29 spurcata, 176, 235 zebrina, Solpuga, 18, 43 Zeriassa, 12, 19, 21, 52, 53 Zeriassa cuneicornis, 52, 53 purcelli, 52, 53; PI. VII, fig. 36 Zonurus, 239 barbertonensis, 240; PI. Ill breyeri, 239, 241; Pis. I, II capensis, 242 cordylus, 242, 243 giganteus, 240, 242 jonesi, 243 robertsi, 241; PI. IV tropidogaster, 243 vittifer, 242, 243 warreni, 241 Zophodiopsis, 233 hyaenella, 233 PRINTED BY WALTER LEWIS, M.A., AT THE UNIVERSITY PRESS, CAMBRIDGE ,_7,£ ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM VOLUME VII ■'■M 'i- PART i containing A Short Survey of the Solifugae of South Africa. By John Hewitt. (With 8 plates and 14 text figures) Addendum. (With 1 text figure) Issued 31 st October , 1919 PRINTED AT THE UNIVERSITY PRESS CAMBRIDGE, ENGLAND i9r9 n £ ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM Vol. 7 Part i A SHORT SURVEY OF THE SOLIFUGAE OF SOUTH AFRICA By John Hewitt With 8 plates and 14 text figures CONTENTS PAGE Introduction ........... 3 General Remarks on the external structure of SOLIFUGAE . . 4 External sexual differences ......... 8 Notes on specific characters . . . . . . . . 12 Distribution of Species ......... 16 Key to the Genera of S. African SOLIFUGAE . . . . . 21 Genus Solpuga Licht. .......... 24 Key to the S. African species of the Genus Solpuga .... 43 Genus Zeriassa Pocock . . . . . . 52 Key to the S. African species of Zeriassa . . . . . . 53 Genus Daesia Karsch . • . . . . . . . 53 Key to the S. African species of Daesia . . . . <» 55 Genus Blossia Simon ........ * * 56 Key to S. African species of Blossia ....... 59 1 2 Annals of the Transvaal Museum PAGE Genus Gluviopsis Kraepelin .......... 61 Genus Hemiblossia Kraepelin . . . . . . . .61 Key to Species of Hemiblossia . . . . . . . 62 Genus Melanoblossia Purcell ........ 62 Key to Species of Melanoblossia ..... . . 62 Genus Toreus Purcell ......... 63 Genus Ceroma K. .......... 63 Key to Ceroma ........... 64 Genus Lipophaga Purcell ....... 65 Key to Lipophaga ........ 65 Genus Hexisopus Karsch . . . . . . . . . 66 Key to Hexisopus ......... 69 Genus Chelypus Purcell . . . . . . • • 7° Key to Chelypus ......... 72 Addendum Solpuga ....... 73 Explanation of Plates . . . . . . • • • 74 List of Recent literature 75 A SHORT SURVEY OF THE SOLIFUGAE OF SOUTH AFRICA INTRODUCTION. The study of the Solifugae has absorbed a certain amount of attention from Zoologists for many years. Morphologists regard them with interest because of the retention of primitive characters in their structure, as is conspicuously exhibited in the segmentation of the body: the sys- tematise on the other hand, is equally impressed by the remarkable diversity of forms found within the same genus, the variations of which seem to be promiscuous and largely discontinuous. The Solifuge fauna of S. Africa is particularly rich: indeed, a fauna so abounding in genera and species as that found in the western and central portions of this subcontinent is not known from any other part of the world. For our knowledge of the species we are indebted to various workers, but more particularly to Mr R. I. Pocock, Prof. K. Kraepelin and Dr W. F. Purcell. Mr Pocock’s pioneer papers on the material in the British Museum of Natural History constitute the first important attempt to make known the great diversity that obtains within the limits of this order, and to classify the genera and species. His work considerably advanced the knowledge of this subject, but was somewhat marred by the very inferior illustrations that accompanied his accounts. A few years later, Prof. Kraepelin’s very useful monograph of the whole order was published in Das Tierreich. In this work all the known species were described, and the important characters as far as possible illustrated. Thus, it was, and still is, quite indispensable to students of this order, but is no longer sufficient as a guide to the species of the less familiar genera such as Blossia. The best contributions to our knowledge of S. African Solifugae are those contained in Dr Purcell’s several papers, based on the collections of the S. African Museum. These are sufficiently accurate and detailed to be accepted as a very reliable starting point in the study of our fauna. Since the appearance of the above mentioned works, various new species have been described by the late Prof. Kraepelin and by the present writer. Kraepelin’s papers are useful as presenting his final accounts of the fauna of South-West Africa and of the Kalahari. During recent years a great deal of additional material has accumu- lated in the collections of the Museums of S. Africa, and although ade- quate material for even a moderately complete account of the variation exhibited within this order in S. Africa is still unavailable, yet I think a sufficient amount of new data has been obtained to justify the following revision of all the known species and varieties. The imperfections of our knowledge will be understood from the fact that very many species are only known from single specimens. The present paper, based mainly on the collections of the Albany and i — 2 4 Annals of the Transvaal Museum Transvaal Museums, is presented as a guide to the characters and gross distribution of the various genera and species now known to inhabit South Africa, the northern limits of which are the Zambesi and Cunene rivers on the east and west sides respectively: all the distribution data of earlier authors have been incorporated, but for detailed descriptions of the various species reference must be made to the original accounts. SOME GENERAL REMARKS ON THE EXTERNAL STRUCTURE OF SOLIFUGAE. Some of the external characters presented by the Solifugae are very stable: the segmentation of the body and of its appendages (excluding the tarsi of the legs), and the position of the various apertures of the body, are thus not available for systematic purposes. The most important characters employed in the classification of the S. African forms are: The segmentation of the tarsi of the legs, the structure of the male flagellum, the nature of the dentition, and the modification of the hairs and bristles which clothe the various surfaces. Most workers have also attached considerable importance to the features of the very character- istic rostrum or camerostome1 which has the mouth at its apex, but such variations as do occur in this structure amongst the various genera are not very striking, and seem to me of doubtful value in a natural classifi- cation. The segmentation of the body is described in a general way in most text books of invertebrate zoology, and can be easily determined from fresh specimens or spirit-preserved material. Some confusion may arise in the case of Hexisopus and Chelypus where the tergites of the abdomen are not strongly chitinised, and thus are not easily distinguished. A source of difficulty may also be found in the segmentation of the thorax, the dorsal sclerites of which are much reduced in all Solifugae. In the accompanying figures, the relationship of these sclerites to the appen- dages has been indicated in accordance with the views expressed by Mr H. M. Bernard in his important treatise on the morphology of this order (27). It should be mentioned, however, that the celebrated arachnologist W. Sorensen has recently presented a different interpretation of the sclerites behind the head-plate (28). He recognises only two thoracic tergites, viz. those labelled as third and fourth in this paper: according to him, those here labelled as first and second cannot be true tergites, because the elevator muscles of the appendages are not inserted thereon. Sorensen thus interprets the structure of a solifuge: head bearing four 1 This has been homologised by Croneberg and by Gaskell (see The Origin of Vertebrates, pp. 222, 223) with the first antennae of Crustaceans: for another view see Bernard’s monograph. The terms applied in this paper to the various appendages are those in current use amongst Arachnologists, and have no reference to their respective homologies with the appendages of other Arthropods. I may remark that the very characteristic chelicerae — sometimes unfortunately termed the mandibles — are apparently homologous with the second antennae of Crustaceans and with the antennae of insects, whilst the pedipalpi or palps are homologous with the mandibles, of other Arthropods. Annals of the Transvaal Museum 5 pairs of appendages, thorax with two tergites and two pairs of appen- dages, abdomen of u segments, the first of which is greatly reduced whilst the second or genital segment is largest. With regard to the thoracic tergites, I prefer Bernard’s view seeing that the bristly arma- ment of all four sclerites (the second excepted) in Blossia has much in common but differs from that of the head-plate : in short, they seem to be homologous structures. The two separated portions of the second tergite are certainly devoid of spines or bristles in Blossia, but a few setae occur near the anterior edge in Hemiblossia. It may be added Text fig. i. Dorsal and lateral views of the cephalothoracic region of a female Blossia sp. that the third tergite, though not divided, has its spines in two separated groups. For an ingenious hypothetical explanation of the composition of the head-plate, the reader is also referred to Mr Bernard’s paper. Here, I only mention this structure in order to direct attention to variations therein. Amongst the various S. African genera there are considerable differences in the shape of the head-plate, the proportions of which more- over may differ greatly in the two sexes of the same species, the female being more robust than the male. In most genera, the lateral portion, 6 Annals of the Transvaal Museum extending ventralwards on each side, is not very extensive, but in Chelypus and Hexisopus the lateral extensions are quite strongly de- veloped, and the head-region is thus particularly robust. Normally, the lower portion of this lateral extension is separated from the upper only by a deep broad groove, but in Blossia the compound nature of the structure is indicated by a well-marked suture line, running immediately ventral to the somewhat ill-defined groove and separating off the lowest portion of the lateral extension as a distinct sclerite: the suture line is faintly indicated also in other genera. This sclerite may possibly be an isolated anterior portion of the first tergite, but in Blossia is not com- pletely fused with the forward continuation of what appears to be the first thoracic tergite. It may also be noticed that whereas in most Solifugae the anterior lateral lobe of the head-plate, which is said to bear rudimentary eyes1, is only partially separated from the main plate, a more complete separation occurs in Chelypus, where a fairly distinct suture line extends backwards to the posterior margin of the plate. The segmentation of the legs, excluding the tarsi, is fundamentally similar throughout the order. Dr Purcell in describing Hexisopus (9) laid some stress on the supposed fact that this genus has a reduced number of trochanter segments, thus differing from all other known genera. The same view seems to have been held previously by Mr Simon, for, in his account of Hexisopus fodiens, the trochanter segments are given as two, and the tarsal segments also as two. In Das Tierreich, Purcell’s views on the homology of these leg segments are mentioned, but Kraepelin adopts what is undoubtedly a more correct interpretation, without however presenting any reasons for his homologies. In most Solpugids, the distal trochanter segment of legs II-IV is very character- istic, having a dorsal infolding of chitin marked externally by a definite line extending over the length of the segment. In Hexisopus, this line is scarcely noticeable, and as the third trochanter of leg IV is greatty elongated, like a femur, whilst the true femur is abbreviated, a confusion of homology is not surprising. The third trochanter of leg IV in Chelypus, however, shews the dorsal line very distinctly and there can be no doubt of the identity of the segment. The homologies of the leg segments can also be traced quite independently from the character of the articula- tions, certain of which are quite distinctive. In a Solpuga, there are specialised areas of thickened chitin arranged in pairs at particular joints, where the movements of the segments concerned are restricted to one plane, viz. between the patella and tibia of the palp, between the femur and patella, and between the patella and tibia of all the legs. In the second or third leg of Chelypus these can be seen without difficulty, and, relying on this character alone, the third leg of Chelypus is found to have three fairly large trochanter segments but only one tarsal segment. In the terminology of the segments of the legs, I prefer to follow 1 W. Sorensen was unable to find the rudimentary eyes, nor could the present writer. In the recent edition of the Encyclopaedia Bvitannica, E. R. Lankester speaks of “a pair of median eyes and obsolete lateral eyes on each side.” Annals of the Transvaal Museum 7 Pocock’s scheme1, as given in the Arachnida volume of the Fauna of British India, rather than that adopted by Kraepelin or the slightly different one of Sorensen: it may be noted, however, that Pocock’s account is not free from error, for he represents legs II-IV as having each the same number of trochanter segments. The segments of the palp are: coxa, trochanter, femur, patella, tibia, and tarsus, which latter is usually fixed immovably to the tibia, is without claws, and is composed of one segment, excluding the rudimentary segment or seg- ments found in the terminal sucker, which according to Sorensen is morphologically equivalent to the claw-bearing plantula of a walking leg. The segments of the legs are similar, but between the coxa and the femur there are several trochanter segments, viz. two in legs I and II, and three in legs III and IV : a special term is given to the distal trochan- ter segment in each case, viz. the trochantin, although it is represented both by Kraepelin and Sorensen as “the basal joint of the femur/ ’ and thus equivalent to the single segment called trochanter in the palp. The middle trochanter segment of legs III and IV is termed the tro- chantella. This and the proximal segment to which the term trochanter now becomes restricted are, according to Sorensen, parts of the coxa — a conclusion which seems to me very reasonable from consideration of such a case as the fourth leg of Chelypus. The joints between trochan- tella and trochantin, and between trochantin and femur, permit of a good deal of twisting of the leg, but, as previously mentioned, the next two joints only permit of movements in one plane. The tarsi of the legs present a character which varies considerably throughout the order, although within the limits of the same genus the tarsal characters are generally very constant in specimens of all ages. The segmentation of the tarsi is utilised as a very convenient generic character, the range of which is sufficiently indicated in my key to the genera. In various genera, the tarsus of the fourth leg is more numerously segmented than the tarsi of preceding legs, and as this multi-segmented condition is presumably secondary, the occurrence of minor segmented aberrations from the normal may perhaps be expected. I believe that such aberrations will prove to be not very uncommon. One such has even been made the type of a distinct genus (Broomiella Pock.), for this seems to be founded on an abnormal specimen of Daesia sehreineri, having a two-jointed fourth tarsus instead of the usual four-jointed tarsus. I have also seen a male of Solpuga hostilis from Doornkop, in 1 Nevertheless, within the comparatively narrow Lmits of the Arachnida, the homologies of the individual segments of the legs amongst the various orders are by no means certain : whiJst a uniform nomenclature for the Arthropoda as a whole seems quite impracticable on the basis of homology. Many morphologists believe that a. though the palp in many orders is 6-jointed, yet the segmentation is not strictly homologous throughout as Pocock on the other hand has represented it. This view is set forth in Simon’s important work Hisioive natuvelle des Avaignees. Thus, a patella of the type found in spiders is held to be wanting in the Solifugae, and it must be admitted that the nature of the articulations is quite different in these two orders: the segment termed patella in this paper is represented to be equivalent to the combined patella and tibia of spiders, and the segment here called tibia as homologous with the metatarsus of spiders. 8 Annals of the Transvaal Museum which the third leg on the right side has a seven-jointed tarsus, like the fourth leg, whilst that on the left side is four-jointed as usual: the right third leg is apparently not abnormal in any other way, the basal portions being like those of a normal third leg. The spinulation of the legs is also a very constant character within the same genus. Spines are most strongly developed in the genus Solpuga, where legs III and IV are spined as follows: Fourth leg, patella with a pair interiorly near the apex, tibia with five (sometimes six or four) on the anterior surface, and three (but varying from two to five) on the posterior surface, tarsus with ten pairs interiorly (or ii-io) : Third leg, patella with two at the apex superiorly and three on the inferior surface, including a pair at the apex, tibia with five superiorly and three pairs inferiorly (occasionally 4-3 or 4-4), tarsus with seven pairs inferiorly. The first leg is completely devoid of spines, and the second leg is spined more or less like the third but the tibia may have 3*2 or 2-i spines ventrally and the tarsus four spines or none externally above. These are all strong spines : in addition, there may be shorter and weaker ones on the second and third tarsi inferiorly. In all other genera the number of spines on the tarsi are fewer: in Daesia, the tarsus of leg IV has three pairs of spines infero-laterally. EXTERNAL SEXUAL DIFFERENCES. Externally, the sexes are distinguished primarily on the characters of the first abdominal sternite (it is actually the second sternite according to Sorensen). This genital sternite is seen in its simplest form in the adult female of the genus Lipophaga, where the two flaps, representing a pair of appendages, remain quite distinct from each other and either one can be raised independently: the posterior mesial angle of each flap is acute. In other genera, these flaps are firmly united along the midline by membrane: sometimes, as in Solpuga Venator, union takes place along the whole length of the flaps, and their hind borders are broadly rounded on each side : occasionally, as in Solpuga chelicornis and hostilis, an acute posterior lobe at the apex of each flap remains free, a condition which presumably is more primitive than that of Venator. The female genital aperture lies in the soft skin posterior to and protected by the sternite, and is quite large in adults. In adult females of Solpuga there is often a small brown scar or several scars on the anterior portion of the genital sternite mesially: this is presumably a mark made by the male during the mating process and is thus indicative of sexual maturity1. On the other hand, the genital aperture of the adult male opens on the surface of the same sternite. The two halves are united together, but in the middle they enclose a pair of long convexly raised sclerites between which mesially is the elongated slit-like genital aperture. These sclerites occur in young males as well as in adults but are much larger in the adults relatively as well as absolutely. 1 An account of the mating habits of Galeodes is given by R. Heymons in a paper entitled ‘ ' Biologische Beobachtungen an asiatischen Solifugen” in Abh. Preuss. Akad. Wiss. Berlin, 1901. Annals of the Transvaal Museum 9 Each half of the genital sternite is actually a compound structure in both sexes : the parts of which it is composed are entirely fused together posteriorly, but anteriorly the suture lines can be easily re- cognised. The half sternite is therefore in- terpreted by Bernard as derived from two segments of an appendage incompletely separated by interarticular membrane, and I may remark that the so-called segmental and interarticular regions are somewhat different in their coverings of hair : in the example figured (. Daesia lineata) the tri- angular interarticular portion is clothed only with comparatively short fine hairs whilst the rest of the sternite bears also a great number of much longer and stiffer setae which are cleft or even trifurcated at the tips. In most species, the adult male is characterised by the possession of the flagellum on each chelicera. This organ is derived from a socketed bristle, and thus the rotatable flagellum of a Blossia presumably repre- sents a more primitive condition than that of Solpuga which is fixed. Dr Purcell has described an enlarged feather bristle in the genus Melano- blossia as a flagellum, and in other genera could find no flagellum what- ever in what appeared to be the adult male. It now appears that more or less enlarged feather bristles may co-exist with a true flagellum in the genus Blossia: this occurs in the species B. falcicornis and B. filicornis, where some of the more distal feathered bristles of the series on the mesial surface of the chelicera are considerably longer and stronger than those proximally situated. Moreover, the position of the point of attachment of the flagellum relative to the series of feather bristles is very variable : sometimes in the genus Blossia it is at the distal end of the line of feather bristles, but in Blossia falcifera it lies between the distal enlarged bristles and the dental series, whilst in the genus Daesia the base of the flagellum is far removed from the line of feather bristles. However, Sorensen, after examining in some detail the structure and homology of the flagella of several genera, concludes that morpho- logically the flagellum is the superior bristle or the two superior bristles of the series. I am satisfied that Sorensen's conclusion is quite correct. Not only is the original relation to the line of feather bristles retained in the genus Blossia, but here too the flagellum is primitive in form, being a greatly inflated bristle cut open along its length. The genus Solpuga, which is far more specialised, affords confirmatory evidence, although the flagel- lum of adults has little resemblance to a bristle. The mesial surface of the upper j aw of a very young Solpuga presents two long oblique series of bristles, that adjacent to the cutting edge including about 23 bristles, all feathered with the exception of the distal one which is fairly long but simple : the other series, parallel thereto, consists entirely of simple Text fig. 1 a. Genital sternite in the adult female of Daesia lineata'. flattened out. That of the male is similar, but has also a pair of elongated scle- rites mesially. 10 Annals of the Transvaal Museum bristles, the basal ones stout and strong and the distal ones much more slender. In older specimens, the feather bristles are more numerous and may occupy several rows. On examining a subadult male of S. derbiana, as yet without a flagellum proper, I find near the distal end of the feathered series several simple bristles of which one is markedly stouter than the rest and strongly curved like the feather bristles. This enlarged bristle, which seems to correspond with the single one at the distal end of the feathered series of juvenile specimens, is presumably destined to become the flagellum at the last moult. In such genera as Blossia and Hemiblossia the feather bristles of the adult, in both sexes, occupy only a single series of about 14-17 : they are all feathered, the distal ones often larger than, but not so strongly feathered as, the basal bristles. Daesia has more numerous bristles arranged in a double row in the adult, all being feathered. The primitive flagellum of the Daesiinae is simply a membrane with more or less infolded edges which basally unite to form a cup : the rota- table flagellum of Hexisopus and of Ceroma is probably to be derived therefrom by more extensive fusion throughout its length, whereby the free membrane becomes converted into a flattened tubular shaft: the fixed flagellum of Solpuga has a swollen basal enlargement which prob- ably corresponds to the cup-like base found in Blossia, and the more or less elongated shaft is usually if not always perforated by a fine canal along its length. The flagellum of Solpuga is far more variable in form than the primitive flagella of the Daesiinae and Hexisopodinae : in the least modified species such as suffusca, it lies on the flat or convex upper surface of the chelicera, but in various specialised forms of the hostilis group the basal portions of the flagellum become sunk into a distinct depression of the inner and upper surfaces of the jaw. In the Daesiinae the adult males are often provided with curiously modified bristles on the second abdominal sternite: these are absent in females or if present are not so highly developed as in males. When bristles and spines occur over the surfaces of the appendages and body they are more strongly developed in males than in females, except on the mesial surfaces of the jaws. Other secondary sexual characters are presented by the dentition, which is often greatly modified in the adult male but primitive in the female : the malleoli of males are much larger than those of the females : lastly, males are more slender than females, having longer limbs and smaller bodies but the disparity in size is not great. The adult male of Solpuga has an organ on the palp, viz. the Scopula, which is quite absent in females:. It is composed of numerous, closely packed, short, feathered hairs, which are presumably sensory in function. It is interesting to notice that a similar organ occurs on the tarsi of the legs in the adult male of Stasimopus (trap-door spider), but not in the female. I have elsewhere1 presented reasons for regarding the Scopula of Stasimopus as an organ of smell. There is an important difference, 1 "Note on the occurrence of a pedal nose in the male of a trap-door spider (Stasimopus)” in South African Journal of Science, March, 1917. Annals of the Transvaal Museum ii Text fig. i b. A. Mesial surface of upper jaw of young Solpuga from Pretoria (? globicornis), having only 3 malleoli : shewing parallel rows of feather bristles and of sinple bristles. B. Distal portion of upper jaw of subadult male of Solpuga derbiana (basal portions only of some of the simple bristles are indicated). C. Distal portion of upper jaw of adult male of Solpuga derbiana. [/ = flagellar bristle, b = point of origin of flagellum.] 12 Annals of the Transvaal Museum however, in the hairs of the two Scopulae: those of Stasimopus are simple, but truncated or more or less trumpet-shaped at the tips where the protoplasmic core comes into close contact with the exterior: those of Solpuga are well feathered, and the tips finely pointed. If the receptive portion of an olfactory organ must necessarily permit of direct contact between living protoplasm and odoriferous particles, it does not seem probable that the scopula of Solpuga can subserve an olfactory function. Perhaps the truncated bristles such as occur commonly on the palps almost throughout the Solifugae will prove to be olfactory. Although I have not specially searched for sensory organs I may add that organs, apparently of sensory function, have been found by H. J. Hansen1 and by H. M. Bernard and were identified as Lyriform organs. In the Solifugae they are almost confined to the chelicerae. In addition, there are what seem to be sensory organs near to the cutting edges of the jaws : over the surface of the fang and on the outer side of the jaw near to the row of teeth the thick chitin is pierced by numerous very long canals, each opening by a small pore on the surface. Similar organs occur in great numbers on the legs of various spiders, along with the several more complicated structures to which the term " lyriform” was originally applied. NOTES ON SPECIFIC CHARACTERS. The characters employed in the discrimination of species are primarily those peculiar to the adult male. Although various authors have also attempted to distinguish species on female characters, and have drawn up keys to the species of Solpuga based mainly on such characters as are afforded by measurements and colour, yet with the larger amount of material now available it seems to me quite impossible in many cases to distinguish between the females of closely allied species. They are often much more generalised than the males. This is markedly the case in respect to the dentition, which is highly variable amongst the males of different species and genera, but is essentially identical through- out the females of many species of Solpuga and even of widely different genera such as Zeriassa, Blossia and Daesia. This type of dentition, found both in males and females of Solpuga Venator or of S. lethalis, is without doubt primitive. On the dental characters, the females of the genus Solpuga can be divided into several groups, the largest including all those species ex- hibiting the primitive type of dentition; one small aberrant group includes S. lineata and allies, which have a much modified dentition in the upper jaw of both sexes, and might with some propriety be assigned the rank of a distinct genus: the two species picta and schnitzel, only known from female specimens, constitute a third group; another little group is that of 5. hastata and allies, and lastly the species 5. fusca differs from all others in the genus in the character of two or three inter- mediate teeth, instead of one only, in the lower jaw. On the other characters found in females, it is possible to divide the genus a little 1 “Organs and characters in different Orders of Arachnids.” Ent. Med. u. a. Ent. For. Fr. Meinert, 1893, p. 178, Kjobenhavn. Annals of the Transvaal Museum 13 further. These characters are: the colouration, which does not vary much within a species, being similar in the two sexes, and fairly constant throughout large groups of species : the shape of the hind borders of the genital sternite, whether broadly rounded on each side or produced into a pair of mesial lobes, the former condition being correlated on the whole with plain colouration, and the latter condition with strongly contrasting colours such as occur in the hostilis group: also, but of somewhat less value is the degree of hairiness of the hind legs, long silky hairs being most strikingly developed in 5. sericea, whilst such hairs are absent in Venator and allies but more or less strongly developed in the species of the hostilis group: finally, the proportions in length of the segments of the palps, and legs, compared together and with the breadth of the head-plate, and the degree of development of cylinder bristles on the lower surfaces of the palp are characters which may serve to distinguish the females of certain species from each other. But, having thus rele- gated a specimen to its natural group, it is usually impossible to proceed further. Dr Purcell has commented on the difficulty of distinguishing between the adult females of 5. Venator and S. lethalis, and I may add that the three species ferox, schonlandi and globicornis, though very markedly distinct in the male sex, are practically identical in females. In this paper I have therefore largely ignored the females when distin- guishing between species, and those species which have been founded only on female specimens by previous authors are now placed aside as incertae sedis : most of them can never be identified, except through the locality data. The systematics of the genus Solpuga is largely a study of variation in the male flagellum. This organ presents quite a bewildering series of forms throughout the genus, but usually shews a high degree of constancy in the same species. Many of its forms shew no obvious relationship to each other: indeed, to a large extent the variation appears to be quite erratic and discontinuous. Species which in structure are practically identical as females, and which are no doubt closely related, may never- theless have profoundly different flagella, as is the case in ferox , schon- landi and globicornis, the flagella of which so far as is known are not con- nected together, even indirectly through other species. It is obvious therefore that as a guide to genetic affinity the characters of this organ may be of very subordinate value. Discontinuous variation is noticeable in every section of the genus, but is accompanied by considerable varia- tion of a continuous type. Dr Purcell has remarked on the variation in length exhibited by the flagellum of 5. Venator in specimens from various localities (9). In this case, a fairly wide range is observed, and it might be possible to distinguish several races or local forms on differences in the length of the shaft, the northern varieties having a longer flagellum than southern forms. In such a continuous series, the elimination of the intermediate forms would result in the formation of quite distinct groups. However, I do not know of any pair of species which differ from each other only in the length of the flagellum, and as a matter of fact, the observed variation in length of the flagellum in most species is limited to a comparatively narrow range. Nevertheless, in some sections of the 14 Annals of the Transvaal Museum genus, the characters which distinguish the several forms commonly termed species are not of an essentially discontinuous type. There is a small group of species, characterised by the more or less strongly twisted shaft of the flagellum, and the presence of serrated edges along some portion of this shaft. These species ( spiralicornis , serraticornis and strepsiceros ) seem to represent separate links in a chain of continuously varying forms: a complete series of intermediate forms is not yet known, but sufficient variation has been noted in a small series of spiralicornis, and in several specimens of serraticornis, to indicate that the specific distinctions, though greater in magnitude, are essentially of the same kind as those which are now included within the limits of the same species, and which are clearly of the continuous type. I suspect it will eventually be found that the species cervina, collinita and alcicornis are also forms of one continuous series : and apparently another such series is that of lethalis typicus, lethalis rectus and furcifera. Amongst the nocturnal species, which constitute the most primitive group of the genus, it sometimes happens that the only noticeable differences between species are those of the flagellum. This is the case with S. lethalis and S. Venator, for the minor difference of dentition which is also said to distinguish them is not constant, and moreover is com- monly found within the limits of a single species. In these nocturnal species, the dentition of the male greatly resembles that of the female, and several other characters of the chelicerae, viz. the strong development of stridulatorv ridges, and the abundance of well-feathered bristles in both upper and lower jaws are common to both sexes. More usually, profound differences of dentition, or of spinulation, accompany the variations of the flagellum. Sometimes indeed, the characters of the flagellum may remain very constant throughout a group of forms which differ amongst themselves in the dental characters. This is the case, at least so far as the shaft of the flagellum is concerned, throughout the species hostilis, derbiana, and tookei, the most character- istic feature of the shaft being its sharp blade-like termination : this same blade with modifications also occurs in hamata and bechuanica : it seems to be quite constant in hostilis, but is variable in derbiana, and therefore cannot be regarded as an absolute unit character in a strict sense. The variations of dentition in the hostilis group of species are indeed very numerous. The dentition is comparatively constant within the limits of any one form, and such forms as hostilis may have a fairly wide geographical range : but, we are still uncertain whether the various types are fundamentally distinct, or are units of one or several continuous series. The occurrence of a dentition so aberrant as that of junodi in the midst of an area occupied by allies ( hostilis , marshalli) which share the most characteristic feature of quite a different dental type, is suggestive of mutational variation. On the other hand, the additional material received during recent years has to some extent served to bridge the wide gaps which formerly seemed to separate types so distinct as chelicornis, hostilis and junodi. The typical form of chelicornis, found in the karroid portions of the Cape, is represented at Kakamas by a dis- Annals of the Transvaal Museum i5 tinct variety, macrognathus, with elongated jaws, which, in the dentition, is not very different from hamata found in the Waterberg district; and again, either of the two latter forms can by slight modification be changed into bechuanica, as found at Serowe in the Bechuanaland pro- tectorate ; this by reduction of the two anterior teeth leads on to hostilis of the Transvaal and marshalli of Mashonaland, or by enlargement of the same teeth to junodi of the Zoutpansberg and Waterberg districts. The known facts seem to be easily interpreted as the results of continuous variation, but it is proper to add that a complete series of intermediates is unknown, and that, from the nature of the case, all possible variants of the dentition could be arranged within an apparently continuous series arbitrarily chosen. Other characters of systematic value, found amongst males, are the spines or bristles on the upper and outer surfaces of the chelicerae, and the tooth or keel which is often present on the dorsal edge of the mesial surface of the upper fang. These characters in particular species are often highly developed, and present the appearance of hypertrophied structures. The dorsal tooth of the fang of ferox, for example, is very markedly stronger than that of any other species. This hypertrophied appearance, which is also exhibited by the characters of the flagellum (cp. that of the cervina group) and of the dentition (cp. marshalli) in various species, seems to suggest that varia- tion, either continuous or discontinuous, has proceeded uncontrolled beyond the limits actually required by the creature for the maintenance of its race. That is to say, it seems improbable that natural selection can have been the sole guiding factor in directing the course of variation. Unfortunately, this view cannot be checked by an adequate body of facts based on acquaintance with the mode of life of the various species. The function of the flagellum itself is unknown: it is not a weapon of offence or defence, and according to Heymons’ account of the breeding habits in Galeodes, is not employed during the mating process. Never- theless, Sorensen believes that the flagellum is eminently adapted to the function of handling spermatophores and states emphatically: “Le flagellum est l’organe copulateur des Solifuges.” At present, this is un- supported by observations on the living animal, and to me the hypo- thesis seems improbable in view of the extraordinary diversity in form presented by the shaft and the complete absence of the flagellum in some genera1. 1 See also R. I. Pocock in A Monograph of the Terrestrial Carboniferous Arachnida of Great Britain, 1911, p. 2: “An important factor in the evolution of terrestrial Arachnida has been, in my opinion, a change from the method of pairing, as practised by Scorpions, to new and special methods, resulting in the modification of a part of one of the prothoracic limbs into an intromittent organ, often of great complexity. This may be seen in the Araneae, where the palps are modified, in the Solifugae, where the mandibles are modified, in some of the Acari, and lastly in the Ricinulei, where the legs of the third pair are modified. Even the Opiliones possess very special secondary reproductive organs.” In the Cambridge edition of the Encyclopaedia Britannica the account of Arachnida by E. Ray Lankester includes the following : “ ? intromittent organ of male (solifuge) lodged on the dorsal side of the first pair of prosomatic appendages.” i6 Annals of the Transvaal Museum In the hostilis group of species, which are all diurnal in habit, it is noteworthy that the stridulatory ridges on the chelicerae are usually reduced or even quite absent in the males, but are well developed in the females. Whilst losing the power to stridulate, — for which, perhaps, they are compensated by increased speed, — the males at the same time add to their ornamental characters: the colours become more vivid, and the fringes of long hair on the legs more strongly developed, the adult male of such species as chelicornis and villosa being quite strikingly handsome. But often, as in derbiana, females are more brilliantly coloured than males. Stridulation is only audible to man in the case of the largest nocturnal species, and perhaps does not operate apart from mastication amongst most solifuges. The nocturnal species shew no sexual differences in colour ornamentation, nor in hair development on the legs. It is in fact the general rule in this genus, that all the species exhibiting specialisations of structure or of colour, in one sex or in both, are diurnal in habit: the nocturnal species are all primitive in structure and plain-coloured. DISTRIBUTION OF SPECIES. Owing to insufficiency of material, the distribution of the species cannot be profitably discussed except in the case of the genus Solpuga. The nocturnal species of Solpuga often have a very wide range of dis- tribution. The common species (S. Venator) of the Karroo and Eastern Province of the Cape extends far into South-West Africa, and another large nocturnal species (S. monteiroi ) in the northern parts of S. Africa seems to extend its range right across the continent from Delagoa to Walfish Bay. The common species of Johannesburg (S. schonlandi) occurs also at Kimberley, and in the Eastern Province of the Cape. The diurnal species on the other hand have usually a very limited distribution. Two diurnal species are only known from the Cape Penin- sula. No one species is known to range from the Transvaal to the Cape, and the species found near Pretoria and Johannesburg (S. hostilis) does not extend so far as Kimberley or Bloemfontein. In Das Tierreich, Kraepelin records S. marshalli from Mashonaland and from Durban, but this is no doubt incorrect. A partial exception to the general rule is found in species which occupy a large area of more or less uniform con- ditions. The handsomely coloured S. chelicornis thus ranges from Nama- qualand to the karroid regions of Eastern Cape Province, though it seems more than likely that this species is not uniform throughout, but composed of a number of structurally distinct forms. Solpuga hostilis , again, enjoys a fairly wide distribution over portions of the high and middle veld of the Transvaal and of adjoining parts in Natal. A still more extensive range has been indicated for 5. sericea Poc., the type of which came from Mashonaland, and which has since been recorded from the Zoutpansberg district by Dr Purcell, and from several localities north of the Zambesi by Mr Hirst : but this case is not so anomalous inasmuch as the species, though diurnal in habit, nevertheless belongs to the large primitive group which includes all the nocturnal species. When the species have been arranged into so many natural groups. Annals of the Transvaal Museum i7 according to their structure, it is of interest to see how those groups are distributed in nature. It is important however to guarantee the accuracy of the scheme as an index to genetic affinity, and for this reason the data presented by the more specialised class of diurnal species can be more safely used than that of the primitive nocturnal species. A large natural group of nine species, including hostilis, marshalli and derbiana , ranges over Southern Rhodesia, Transvaal, Bechuanaland Protectorate, Free State, Natal and Eastern Cape Colony: it does not occur in Western Cape Colony, the western limit, at present known, being at Somerset East. A small group somewhat related to this, but distinctly separated therefrom in structure, includes only the Capetown species S. vincta, the Little Namaqualand species S. spectralis, and an- other western form S. maraisi found at Worcester and Stellenbosch. Another small group, also related to the two just mentioned, includes erythronota of unknown locality, erythronotoides from Victoria West, and lateralis from the districts of Uitenhage, Port Elizabeth, and Bathurst. A fourth group of diurnal species, ranking as a well-marked section, includes the two species chelicornis and villosa, which range throughout the karroid regions of the Cape into Little Namaqualand. In addition, there are two other groups of diurnal species, neither of which is directly related to those already mentioned. One of them in- cludes hastata and two other species which are all from Great Namaqua- land, and no member of the group is known to occur elsewhere. Lastly, the very distinct section comprised by lineata and its ally brevipalpis, occurs widely distributed in the western and central districts of the Cape, excluding the Cape Peninsula: eastwards, its limit appears to be Alice- dale. All these diurnal groups seem to be peculiarly S. African, having, so far as we know, no representatives north of the Zambesi. It will be seen therefore that the subcontinent is thus divided up into so many distinct regions, each of which is the home of one particular natural group and that for the most part these regions do not overlap. An exception to this generalisation is presented in the case of the lineata group, which occupies almost the same region as chelicornis. The rule, however, only applies to groups which are sufficiently closely related: lineata is so remote in structure from chelicornis as almost to warrant generic separation therefrom. It is interesting to notice that these geographical regions coincide — but not rigidly so — with the regions occupied by the natural groups of other animals, belonging even to different phyla of the animal kingdom. The eastern area, occupied by hostilis and its immediate allies, is the same as that occupied by the glabrifrons-latimanus group of the scorpion genus Opisthophthalmus : the nearest allies of this group are: austerus, a Karroo species which agrees fairly with S. chelicornis and villosa in its range: macer , a western species which occurs near Capetown extending northwards as far as Worcester and Ceres, and eastwards for some dis- tance along the Cape Coast : crassimanus which occurs in Little Namaqua- land and Carnarvon, and nitidiceps its close ally which occurs in the Albany, Uitenhage, Somerset East and Cradock districts. The almost universal change of fauna experienced in passing through the Cape 2 i8 Annals of the Transvaal Museum Province from east to west is exhibited in more simple fashion by the water frogs: the species found at Grahamstown is Rana angolensis, the same as occurs throughout the Transvaal, Natal, and Rhodesia: quite a different species occurs at Capetown, viz. Rana fuscigula, which is characteristic of the western half of Cape Province. Again, just as each group of species has its own particular area of distribution, so also each individual species seems to have a special part of each large area for its sole occupation. So far as is known, two diurnal species of Solpuga never occur together, except in the case of species which are very distantly related. This fact may perhaps be held to witness against the mutation hypothesis of species formation so far as this group is concerned. Amongst the numerous more primitive species of Solpuga, only a few natural groups can be recognised with certainty. The northern section of black-striped forms includes sericea of Mashonaland, Zoutpans- berg district, andN. Rhodesia, celeripes of Salisbury, striata of Damaraland and zebrina of the Taru desert in British E. Africa. The well-marked section comprising cervina, collinita, and alcicornis, belongs to the western half of the subcontinent, being known from Clanwilliam, Namaqualand, Willowmore, Keetmanshoop and Kuruman. An equally distinct group is that of strepsiceros, spiralicornis, and serraticornis which occurs in S. Rhodesia, Zoutpansberg and Barberton districts; but, a near relative of serraticornis is the species 5. schlechteri , found in Bushmanland and Great Namaqualand: other species apparently referable to this section range far into tropical Africa, one of them keyserlingi (perhaps a synonym of schweinfurthi) being known to me from N. Nigeria. The group including schonlandi , ferox, globicornis , and sagittaria, which is doubtfully natural, seems to range almost over the same area as that occupied by the hostilis-derbiana group but it has relationships with central African species, for S. niassa seems to be an ally of schonlandi. The species Venator, lethalis, lethalis rectus, d,ndfurcifera, are undoubtedly closely related and belong essentially to the western portion of the sub- continent, Venator extending into the karroid region of the Cape. The precise relationships of the other species, and the inter-relationships of the above groups, are too uncertain to justify any general conclusions therefrom. The main facts concerning the distribution of the more familiar genera are given by Kraepelin in Das Tierreich. Since the publication of that work, the following new genera have been described from S. Africa by Dr Purcell: Melanoblossia, Lipophaga, Toreus, and Chelypus. So far as we know, they are all restricted to Southern Africa, as also is the genus Hexisopus : further, all five genera belong essentially to the western region, which includes also the Karroo and Kalahari. At present, not a single species of any of these genera is known to occur in Natal, Transvaal, Free State, or in east Cape Colony Hemiblossia has been recorded by Pocock from Guatemala, but other- wise is only known from S. Africa: this distribution points to great antiquity for the genus, and is in accordance with its primitive nature. Annals of the Transvaal Museum 19 Of the other genera found in S. Africa, the most widely distributed seems to be Gluviopsis, which occurs as far north as Algiers, and extends into Asia Minor, Arabia, Socotra and Western India. Daesia also occurs throughout Africa (excluding West Africa, from the Congo basin north- wards through the coastal countries as far as Morocco, which regions are all very poor in Solifugae) and extends into Palestine and Arabia. Solpuga ranges throughout Africa, and Blossia seems to have the same range: Zeriassa is only known from the tropical parts of South Africa, and from East Africa as far north as Somaliland. Ceroma extends from the western region of S. Africa, northwards into East Africa. Only one genus, Solpuga, is known to me from Natal: however Kraepelin has recorded a Blossia from “ Moritzburg” which, as Pocock suggests, may mean Pietermaritzburg. The distribution of Solifugae is largely determined by conditions of humidity. They avoid dense forest, but are common in arid regions, and in open bush country, such as that of the Kalahari. I am informed by Mr E. C. Wilmot, who writes from Tsessebe, that a species of Solpuga (probably monteiroi ) may often be seen in the act of climbing trees in search of Cicadas. In the Albany district, they are particularly abundant at Alicedale, a warm locality in the valley of the Bushman’s River, where scrub and bush prevail. Mr Frank Cruden has kindly furnished me with the following brief notes on several species found in that locality: Hemiblossia O’neili is found on light gravelly soil near the river, on flats broken by occasional thornbushes. They are only seen during the bright sunshine of a summer’s day, and then in considerable numbers: they are extremely active, and not easily caught, as they seek refuge in small holes and crevices on the slightest alarm. Specimens may be captured by spraying them with dilute spirits of wine, when the creatures succumb at once. Blossia unguicornis is never seen in the open during daytime, and when exposed to light immediately seeks cover. They are found under stones on the flats and hill sides. Solpuga lineatus occurs chiefly among the rhenoster bush and other scrub, but not in dense bush. On a summer’s day they may often be seen lying on stones exposed to the sun. They are sometimes found in shallow earth cavities under stones. Solpuga Venator occurs on open flats and in broken scrub land. During daytime it takes cover under large stones, but at night may enter houses. At Grahamstown I have only found one species, 5. derbiana. It occurs on the open flats above the town, frequenting grass veld as well as stony lands covered with rhenoster bush and other short scrub. During winter months these creatures hide below ground: they emerge in numbers during the bright sunny days of October. Rapidity of move- ment is their characteristic habit : they run along the ground : they climb the low bushes with ease, at times swinging from twig to twig like long- limbed apes: they are expert burrowers, using the chelicerae as digging organs. Despite their speed they are said to fall a ready prey to kestrels. 2 — 2 20 Annals of the Transvaal Museum The following notes are taken from a little-known paper entitled ‘'Some Arachnids at Hanover, Cape Colony/' by S. C. Cronwright- Schreiner, in the Popular Science Monthly for December 1902: “If you watch a Solpuga closely, you may see its sides palpitating rapidly, even violently, if you hold it in your hand. Like all active, high strung, quick breathing creatures, the Solifugae perish almost instantaneously when immersed in spirits, while large scorpions and large Harpactirae will live for two or three hours. They are great burrowers, but do not make regular holes apparently, and they lie dormant underground during the winter. They are a feature of the thirsty veld and the blazing sun. The ‘Tommies' along the railway sometimes make one of these creatures fight with a scorpion. They place the combatants in some slippery vessel so that they cannot run out. The scorpion is nearly always much the larger and heavier and has in addition to its long arms and powerful nippers, a deadly sting. Yet it not infrequently happens that the jacht- spinnekop comes off victorious, for it seizes the scorpion in its terrible shears and tears a huge hole in it with a quickness and force against which the scorpion is often powerless. When one first sees a Solpuga on the veld, especially the commonest (S. chelicornis) , one can hardly believe it is not a beautiful karoo flower: touch it, and awa}/ it darts. The Dutch call them Jacht Spinnekoppen or Haar Scheerders. Jacht Spinnekop (hunting spider) is a very appropriate name, for, to the casual observer they resemble spiders, and they are mighty hunters. Haar Scheerder (hair shearer) is even more oppropriate. They are called Haar Scheerders because of their two enormous shears. Many a person believes that, if they get into your hair, you will not get them out again until they have shorn it all oh.'' The S. African solifuge fauna is noteworthy for the abundance of diurnal types. Text-books of Zoology and Natural History describe these creatures as wholly nocturnal, and indeed this seems to be actually the case in other parts of the world. In Dr F. Werner’s paper on, “Scor- pions and Allied annulated spiders of the Anglo-Egyptian Sudan " and in Mr Pocock's account in the Fauna of British India, the whole fauna is represented as nocturnal. The following arrangement of the subfamilies and of the several genera in each subfamily has no claim to phylogenetic importance. The subfamily Solpuginae seems to be quite sharply separated from the other groups and I regard it as the most specialised. The Daesiinae constitute a natural group of which Daesia is probably the most specialised genus : this subfamily presents certain points of affinity with the Galeodidae. The Karschiinae have relationships therewith, yet are well worthy of distinction in my opinion, although Prof. Kraepelin finds difficulty in separating them from the Daesiinae: the subfamily includes the most primitive of S. African Solifugae, and indeed the genus Lipophaga would appear to be the most generalised of all known Solifugae. The very specialised Hexisopodinae are somewhat isolated and their affinities, obscure but it does not seem necessary to separate them as a family distinct from the Karschiinae and Daesiinae. Annals of the Transvaal Museum 21 Key to the Genera of South A frican SOLIFUGAE. A. Fourth leg with a pair of well-developed terminal tarsal claws. These claws are somewhat larger than those on the preceding legs. The femora of the legs are all long, much longer than the trochantin. a. First leg without tarsal claws1. av Tarsus of fourth leg with seven segments, of which the basal one is by far the longest, being quite as long as the next four segments together, the three penultimate segments being the shortest: tarsi of second and third legs with only four segments: subungual appendages of the tarsal claws small and not strongly divaricating. S. /. Solpuginae. Subfamily SOLPUGINAE. 1 . Tarsus of palp immovably attached to the tibia, which segment becomes somewhat narrowed distally and on its inferior surface is armed with numerous short truncated bristles and longer pointed ones but no true spines, also in the male often with a very distinct scopula of short feathered hairs: coxa of palp with an elongated maxillary process projecting anteriorly from the ventral border mesially: ocular tubercle bearing a number of stiff bristles or weak spines, several pairs of approximately equal size being directed anteriorly (in juveniles one pair of forwardly directed spines is noticeably larger than the others) : anterior border of head-plate straight and the mesial longitudinal groove more or less obsolete in adults : the glabrous area on the mesial surface of the basal part of the chelicera bearing a series of parallel stridulatory ridges, which however are occasionally absent in adult males : the dental series of the upper jaw flanked on the inner surface and sometimes almost hidden by a strip of numerous feathered bristles, and parallel thereto but situated at a little distance posteriorly is a row of forwardly directed sharp-pointed stout spines, usually 7-9 in males, but more numerous in females : dentition of upper jaw variable. in the single series, but the lower jaw has always two large teeth with one (rarely two or three) intervening small tooth; surfaces of body hairy, with stiffer setae on the chelicerae and head-plate; in the male long spines often occur on the chelicerae, but never on the tergites: flagellum of adult male with a stiff shaft exhibiting great variety in shape, generally cylindrical or more or less flattened into a ribbon; this arises from the basal enlargement, a hollow closed capsule, usually flat on the mesial side and turgid on its external side, and fixed immovably along its base to the upper or inner surface of the upper jaw: the walls of the capsule are thickened along the dorsal and hind margins, and the cavity is continued as a fine tubule into the procurrent portion of the shaft which remains firmly attached to the surface of the jaw up to the point where the shaft bends upwards. (Species usually of large size.) Solpuga Licht. 2. Similar to Solpuga, but the tibia of the palp in the adult male carries a number of stout spines on its inferior surface as well as a scopula, which is not large: ocular tubercle armed with two semicircular series of stiff bristles, the two largest of which project horizontally forwards. Zeriassa Pocock. 1 According to Kraepelin, rudimentary claws may occur in Gluviopsis. 22 Annals of the Transvaal Museum a2. Tarsus of second and third legs with 1-2 segments, of fourth leg with 1-4 segments. Stridulatory ridges of chelicerae long and well developed. Flagellum of male when present is membranous and expanded, at least in its basal portion, the edges incurved to form an open capsule: at the base where it is attached to the mesial surface of the jaw, there is a ball and socket joint by which it can be movably articulated in a vertical plane around its point of attachment. Anterior border of head-plate not quite straight but more or less curved. S. /. Daesiinae. Subfamily DAESIINAE. 3. Second and third legs with two tarsal segments, fourth leg with four tarsal segments, the basal one of which is subequal to the other three in length, the one next to it being shortest, and the most distal segment longer than the penultimate or the antepenultimate. Tibia of second leg with five dorsal spines, of third leg with three dorsal spines. Tarsus of palp slightly movable. Tibia of palp without cylinder bristles or scopula interiorly but with short spines on each side, three externally and 1—3 internally as well as several longer weak spines and bristles. Maxillary process of coxa of palp very short and rounded. Dentition of female very like that of Solpuga Venator : of male very variable. The surfaces of the body are not spiny but covered with hairs and long setae which may be spiniform on the upper surface of the chelicerae and on the hind border of the head-plate of adult males. Adult male with a rotatable flagellum, and with a group of numerous modified hairs on the second sternite. Species of moderate size and plain colouration, strongly contrasting colours being unknown in this genus. Daesia Karsch. 4. Tarsus of second and third legs composed of only one segment, of fourth leg comparatively long and slender, composed of two segments the distal one of which is much the shorter and is quite freely movable. Tibiae of second and third legs without a dorsal row of spines. Head-plate with a very distinct mesial groove, and armed usually with short spines, prickles, and forked bristles, whilst the upper surfaces of the chelicerae and the tergites, especially those of the thorax, are armed with stiff setae and cylinder bristles the longer ones of which are forked at the tip, or in the male with spines which are often very strong, the first thoracic tergite constituting a narrow spiny collar to the head-plate. Ocular tubercle with two stiff bristles projecting horizontally forwards. Tibia of palp inferiorly with spines and numerous cylinder bristles: short cylinder bristles, or long ones forked at the tip, occur also on the surfaces of the body and appendages generally. Maxillary lobe of coxa of palp fairly long and prominent. Adult male with a membranous rotatable flagellum, and there are several pairs of modified fleshy hairs on the second abdominal sternite in the male and sometimes also in the female. Species of small size and plain colour (except B. obscura). Blossia Simon, 5. Similar to Blossia, but flagellum of male is a flexible densely ha ry rod, not rotatable but directed forwards and hidden between the chelicerae. Patella and tibia of palp without spines inferiorly. Ocular tubercle with a semicircle of fine setae on the mesial side of each eye. Anterior margin of head-plate almost semicircular. Upper jaw of male strongly compressed laterally, forming a vertical lamina except quite at the base. Melanoblossia Purcell. Annals of the Transvaal Museum 23 6. Tarsus of fourth leg comparatively stout, especially in the female, with one or two segments, but in the latter case the two segments are subequal in length and not movably articulated with each other. Patella and tibia of palp without true spines below, the tibia being thickly studded all round with brownish black truncated cylinder bristles. Chelicerae of male without spines above, but with long slender curved bristles. Flagellum of male a large flat rigid membrane rotatable at its narrow end, bearing on its mesial surface a delicate capsule with long slit-like opening, and terminating at its broader end in slight exfoliations, two of which are more or less like short horns: second and third sternites with numerous fleshy hairs. Species of small size and with strongly contrasting colours. Hemiblossia Kraepelin. 7. Similar to Hemiblossia, but patella and tibia of palp in both sexes with stout spines below in a double row. Tarsus of fourth leg comprising only one segment, which is comparatively slender (5-6 times as long as deep). Flagellum membranous, forming an inflated capsule open along its mesial side. Second and third sternites of male each bearing two clusters of several modified fleshy hair structures. Gluviopsis Kraepelin. b. Tarsus of first leg with a pair of small or minute tarsal claws. Flagellum of male when present rotatable (or fixed apparently in some species). S. /. Karschiinae. Subfamily KARSCHIINAE. 8. Tarsi of legs II— IV with two freely movable segments, the basal one five or six times as long as the distal segment, which has very long, strongly diverging, subungual appendages. Tibiae of legs II and III spined dorsally, but leg IV without spines. Tibia of palp more or less cylindrical, not attenuated distally, and without true spines or scopula inferiorly, the patella with five very long bristles on each side inferiorly. Upper surfaces of chelicerae and body without strong spines. Anterior margin of head -plate rounded. Ocular tubercle armed with a number of fine bristles especially in front. Basal enlarge- ment of flagellum small and not very sharply differentiated from the shaft, attached by a fairly large rotating joint (sometimes fixed?) to the inner surface of the jaw, and produced directly upwards into the shaft: basally, it is protected and more or less concealed on the mesial side by a fan -like row of feather bristles or by several spines. Upper lobe of rostrum evenly pointed at the apex, the lower margin being not horizontal but inclined upwards. Coxa of palp with a short maxillary process. Ceroma Karsch. 9. Similar thereto, but adult male without a flagellum. Upper lobe Of rostrum unevenly pointed, the lower margin being horizontal and very slightly concave, and the upper margin curving strongly downwards. Toreus Purcell. 10. All the legs with a single tarsal segment and small pul villus. The two halves of the genital sternite of the female not united mesially. Adult male without a flagellum and no fleshy hairs on the second abdominal sternite. Patella of palp with strong spines inferiorly in the male: tibia of palp with numerous truncated cylinder bristles inferiorly especially in the male. Ocular tubercle with a pair of spines anteriorly. Tibia of leg IV with slender spines inferiorly in the male: of legs II and III with four or five stout dorsal spines also. Lipophaga Purcell. 24 Annals of the Transvaal Museum B. Fourth leg without tarsal claws. All the legs are short, the posterior three pairs very robust, their more distal segments strongly spined. In the fourth leg the combined length of coxa and of the three trochanter segments is about equal to that of the rest of the limb, the femur of this leg being shorter than the trochantin. Surfaces of body and appendages clothed with long silky hairs. Flagellum of male rota- tably attached at the small cup-like basal enlargement to the inner surface of the upper jaw. S. /. Hexisopodinae. Subfamily HEXISOPODINAE. 11. The three distal segments of leg IV are terete, or only slightly com- pressed, and without angular edges. Pedipalps not spined. Stridulatory area of chelicerae with parallel ribs. In females, the mesial surface of the chelicera carries feathered bristles and stout simple bristles, but the males are devoid of feather bristles and have no long simple bristles, though dorsally near the base of the -fang there occurs a dense patch of short spiniform setae. The flagellum of the male is hidden between the mandibles, being attached to the jaw far back, but quite near to the cutting edge and not far from the angle of the jaws. H exisopus Karsch. 12. The three distal segments of the fourth leg are broad and more or less strongly flattened with angular edges, and some of the distal segments of the third and fourth legs have their posterior surfaces hairless and densely covered with short granuliform or dentiform spinules. Distal segments of pedipalps strongly spined. Inner surface of chelicerae with a large smooth area marked with fine furrows, which are sometimes more or less reticulately arranged, and sometimes more or less in longitudinal lines. (Females of this genus unknown.) Chelypus Purcell. Genus Solpuga Licht. Solpuga lethalis C. L. Koch, 1842 [Text fig. 2 a], Kraepelin, in Das TierreicK, p. 56, fig. 14. Purcell, in Annals 5. Af. Mus. 1, p. 405, figs. 19 and 19 a. The form described by Dr Purcell, characterised by a well-marked distal sinus on the shaft of the flagellum, was recorded from the following divisions in Cape Colony: Malmesbury, Robertson, Swellendam, Paarl, Worcester, Clanwilliam and Namaqualand. In this form, now termed 5. lethalis typicus, the shaft is devoid of minute serrations except for an oblique band on the anterior half of the sinus. Kraepelin records the species from various localities in S.W. Africa, viz. Ababis, Okasise, Okahandja, Windhuk, Rehoboth and Spitzkoppe near Keet- manshoop. It is not known from the eastern or central districts of the Cape Province. In the male, the spines on the upper surfaces of the chelicerae are not very stout. Solpuga lethalis C. Koch var. nov. rectus [PI. VII, fig. 37 and Text fig. 2 b\. This name is applied to a form which is chiefly distinguished from typical specimens, as described by Kraepelin and Purcell, in the complete absence of a distal sinus on the recurrent portion of the flagellum. The shaft is long, Annals of the Transvaal Museum 25 extending back well beyond the ocular tubercle, approximately reaching the middle of the head-plate: apically it is bifurcated, the upper portion being longer, quite smooth and tapering to a point, the lower portion being short, blunt, and its surfaces well serrated along the angular margins. The shaft is also serrulated along the slight ridges which occur in its basal half, but in the distal half, where ridges are still more marked, the serrations are absent excepting near the apex: the surfaces are in fact comparatively smooth in the distal half but roughened in the basal half. The anterior bend of the flagellum is approximately midway between the apex of the fang and the first tooth. On the inner upper margin of the fang, near to the anterior bend, there is a small tooth. Measurements. Breadth of head-plate 11, length of patella of palp 17*8, of tibia and tarsus of palp 17-5, of patella of fourth leg 16. Total length of recurrent portion of flagellum 16-75. Text fig. 2, a and b. Solpuga lethalis Koch, a, Distal portion of flagellum of typical form from O’okiep, viewed from the outer side interiorly, b, Ditto of 5. lethalis var. nov. rectus from Windhuk: more enlarged than a. The type of this variety is a single male example from Windhuk (G. A. Thompson), in. the collection of the Transvaal Museum. This form was evidently known to Prof. Kraepelin: in his last paper (3) some reference was made to the variability of the flagellum in this species and Venator, but unfortunately no locality data were given for the varietal forms there mentioned. Solpuga venosa Purcell, 1899. Annals S. Af. Mus. 1, p. 412, figs. 18 and 18 a. Kraepelin, in Das Tierreich, p. 74, fig. 41. The type was taken from a locality about twenty miles east of Pietersburg, Zoutpansberg dist. Solpuga fur cif era Kraepelin, 1899. Das Tierreich, p. 79, fig. 56. Kraepelin cites the following localities in the northern parts of S.W. Africa: Osire, Windhuk, Rehoboth, Walfish Bay. A description of the female is given by the same author (2) . 26 Annals of the Transvaal Museum Solpuga fusca C.L. Koch, 1842 [PI. VI, fig. 26]. Purcell in Annals S. Af. Mus. 1, p. 417, figs. 22 and 22 a. Kraepelin in Das Tierreich, p. 79, fig. 54. This species is recorded by Purcell only from localities in the Cape Penin- sula. It may be noted that Kraepelin’s figure in Das Tierreich does not agree with Dr Purcell’s account in regard to the position of the anterior bend of the flagellum. Solpuga toppini Hirst, 1916. Annals Durban Mus . 1, p. 228, fig. 15. The type is from Ngxwala Hill, Ubombo, Zululand. The Durban Museum has several large females from the same locality which are probably referable to this species. The general colour in spirits is a dirty brown with olivaceous tinge, the chelicerae especially somewhat greenish, tergites all blackish, malleoli dark-edged. The single series of teeth in the upper jaw with two small intermediate teeth. Genital sternite with broadly rounded posterior margins. There is a stronger development of fine hairs on the palp and legs than in females of the venaior group, and the head- plate is rather more hirsute. Measurements: breadth of head-plate 12*5, length of tibia and tarsus of palp 11, of patella of palp 10-3, of patella of fourth leg 10-2, of tibia of fourth leg 9. These females are well separated from those of fusca, to which toppini seems to be related, in that the dentition of the lower jaw is quite normal for the genus. Solpuga Venator Pocock [PI. I, fig. 1 and PI. IV, fig. 12], 1897. Ann. Mag. Nat . Hist. 6, xx, p. 258, fig. 7. Purcell in Annals S. Af. Mus. 1, p. 407 and 11, p. 208. The type of this species came from Kleinpoort in the Eastern Karroo, and specimens from Port Elizabeth were identified therewith by Mr Pocock. Recorded by Purcell from the following divisions in Cape Colony : Namaqua- land, Kenhart, Carnarvon, Victoria West, Middelburg, Beaufort West, Prince Albert, Sutherland and Uitenhage; also from Warmbad in Great Namaland. It is known to me from : Brakkloof near Grahamstown (Mrs G. White) ; Alice (Albany Mus, coll.) ; Grattans, Koonap (E. Bennett); Carlisle Bridge (F. Bowker); Alicedale (F. Cruden); Longhcpe (Miss D. Cotton); De Aar (S. C. Cronwright-Schreiner) ; Griquatown (Dr R. Broom); Aus S. W. A. (female examples in Kimberley Mus.). Kraepelin records it from the following localities in S. W. Africa: Warmbad, Churutabis, Keetmanshoop, Kabus, Kuibis, Luderitzbucht, Gibeon and Rehobo th. The spines on the upper surface of the chelicerae are not so strongly developed as in monteiroi. A large female from Keetmanshoop (G. V. Haagner), in the collection of the Transvaal Museum, has the following measurements : Width of head-plate 15, length of patella of palp 16, of tibia and tarsus of palp 17, of fourth patella 14-75, °f fourth tibia 13-3. This specimen has a single intermediate tooth in the single series of the upper jaw. In a male from Dunbrody the single series of teeth in the upper jaw comprises four main teeth, but on the fourth basally there is a distinct additional tooth: in speci- mens from Redhouse a minute additional tooth occurs between the second and third normal teeth. A large female example from Griquatown sent along with an adult male of Venator has two such intermediate teeth, whereas the male has only one: this female may be referable to lethalis, but the dentition cannot be regarded as an infallible guide in distinguishing these two species. A female from Douglas has a single intermediate tooth; males from the same locality may have one such tooth or two, in which case one of them is very small. Annals of the Transvaal Museum 27 Solpuga monteiroi Pocock, 1895 [PI- VII, fig. 38]. Ann. Mag. Nat. Hist. 6, xvi, p. 87. PI. IV, fig. 6 (figure erroneous). Also Ann. Mag. Nat. Hist. 6, xx. p. 257, fig. 6 on p. 261. The author’s second figure was also incorrect in its representation of the spiny armament of the chelicera. The type came from Delagoa Bay. Kraepelin records it from Walfish Bay, and from other localities in the northern portion of South West Africa, viz. Windhuk, Okahandja, Okakena, Osire, and Grossfontein. I have recorded it from Victoria Falls, and it is also known to me from: Mt Temple, Bechuana- land (T. C. Lanham) ; Rooikranz, Rustenburg dist. (Transvaal Mus.); Vyge- boompoort in Waterberg dist. (G. van Dam); Tsessebe (E. C. Wilmot) ; Kraai Pan (Kimberley Mus.) ; Griffin Mine, Leydsdorp (G. van Dam) ; and Newington in N. E. Transvaal (Dr J. P. Fenoulhet). The distribution of this species appears to agree closely with that of the scorpion Uroplectes vittatus Thor. Specimens from Mt Temple have a straight flagellar shaft, whilst those from Tsessebe have a shallow sinuation thereon at a little distance from the apex. In one example from the former locality, the basal enlargement of the flagellum is not so elongate as in typical examples and the spines on the chelicerae not so strong. Female examples apparently referable to this species are distinguishable from those of Venator, globicornis , and allies, from the fact that the patella and tibia of the fourth leg are subequal in length: moreover the cylinder bristles on the palp are comparatively few, the tarsus being devoid thereof and the femur having only very few such bristles. Solpuga schlechteri Purcell [PI. IV, fig. 15], 1899. Annals S. Af. Mus. 1. p. 41 1, figs. 17 and 17 a. The type came from Naroep in Great Bushmanland. In Das Tierreich, Prof. Kraepelin suggested that this may be cospecific with serraticornis : it is clear however that such is not the case, and in Kraepelin’s most recent list the species is maintained on the evidence of specimens from Luderitzbucht and Kuibis, which differ from Purcell’s description only in possessing a single intermediate tooth, instead of two, in the upper jaw. We have the species from Kakamas (Miss H. C. Olivier), and from Keimoes near Upington (S. M. Gadd). In the Keimoes example there is a single intermediate tooth in the upper jaw : in the Kakamas specimens two such teeth are represented, although the second tooth is either a mere rudiment or only minute. The feather bristles of the upper jaw are not strongly developed: they are comparatively numerous on the lower jaw. Solpuga serraticornis Purcell, 1899. Annals S. Af. Mus. 1. p. 409, fig. 16. The type came from the neighbourhood of Bulawayo. Adult specimens from Bulawayo have been kindly lent to me by Dr G. Arnold. In both sexes the dorsal surfaces are somewhat infuscated, but not deeply so, the pigmentation extending over the soft skin for some little distance lateral to the tergites. In the male the spines on the upper and lateral surfaces of the chelicerae are numerous and long, but not very stout : the upper margin of the basal enlargement of the flagellum is curved, not greatly elongated. In the female there are two intermediate teeth in the single series of the upper jaw: cylinder bristles occur on the patella, tibia, and tarsus of the palp, but not many on the patella. 28 Annals of the Transvaal Museum Measurements of adult female. Breadth of head-plate 14, length of patella of palp 12, of tibia and tarsus of palp 12-5, of patella of fourth leg ii-2, of tibia of fourth leg 10*1. A variety of this species, found at Umtali, was described by me under the name of 5. serraticornis umtalica [Annals Transvaal Mus. iv. p. 162, fig. 24). We also have the species from a locality in the south of the Melsetter dist. (F. A. O. Pym). In umtalica the upper margin of the basal enlargement of the flagellum is curved: there are numerous well-feathered bristles on the lower jaw, and the distance from the tip of the fang to the apex of the first tooth is equal to i^-if times the distance between the two large teeth : there is a dense group of long and strong spines on the upper and outer surfaces of the chelicera. Solpuga spiralicornis Puicell [PI. V, fig. 18], 1903. Novitates Zoologicae, x. p. 304, fig. 1. The type came from Shilowane, near Leysdorp. What seems to be the same species was described by Mr S. Hirst from the Zoutpansberg dist., under the name of S. pugilator {Ann. Mag. Nat. Hist. 8, ix. p. 232). It is known to me from Newington (Dr J. P. Fenoulhet), and the Transvaal Museum has it from Hectorspruit (F. Streeter), and Griffin Mine, Leysdorp (G. van Dam). In this species the form of the basal enlargement varies somewhat. In the Newington specimen the upper margin of the basal enlargement is quite long and practically straight : in the Griffin Mine example it is a little shorter, and distinctly curved on the upper margin. Moreover, in the latter, the terminal portion of the shaft beyond the sinus is not so sharply pointed nor so straight as in our Newington example. The Griffin Mine form is presumably typical: that from Newington and Hectorspruit is apparently the same as pugilator Hirst, which may thus rank provisionally as a varietal form. The characters of the lower jaw, however, seem to separate these forms a little further: in the Newington example, the fang is distinctly longer than in other specimens, the distance from the tip to the apex of the first tooth being equal to 1^ times the distance between the apices of the two large teeth, whereas in examples from Griffin Mine and Hectorspruit the proportion is if times. In each case there are numerous well-feathered bristles on the lower jaw. A female example probably referable to spiralicornis was also taken at Griffin Mine by Mr van Dam : I refer it to this species, rather than to monteiroi, which was taken in the same neighbourhood, on account of the dark pigmen- tation on the abdomen, a character found in the male of spiralicornis. The soft skin between the tergites of the abdomen and thorax is rather deeply infuscated, and on the abdomen this extends to some extent down the sides in the neighbourhood of the first three or four segments : in the hinder half of the abdomen the soft skin of the sides is deeply infuscated over an extensive area and also ventrally between the sternites. The head-plate is brown, dark brown near the anterior margin, the ocular tubercle very darkly so. Palps and legs pale yellowish brown. Malleoli not infuscated. There are two inter- mediate teeth of moderate size in the single series of the upper jaw. Measurements. Breadth of head-plate 9*25, length of tibia and tarsus of palp 11, of patella of palp 10-35, °f tibia of fourth leg 9-3, of patella of fourth leg 9-7. Solpuga strepsiceros Kraepelin, 1899 [PI. IV, fig. 13 and Text-fig. 3]. Das Tierreich, p. 68, fig. 31. Annals of the Transvaal Museum 29 The locality cited by Kraepelin is Delagoa Bay. The species is known to me from Barberton (Transvaal Mus.). In Barberton specimens, the flagellum has a spiral twist, but the spiral is not an open one, and the curves not so pronounced as represented in Kraepelin’s figure in Das Tierreich. The serrated edge is not continued over the whole length of the shaft, but commences slightly posterior to the hind margin of the basal enlargement. Basal enlargement short, with curved upper margin. On the upper surface of the chelicera there are a few long slender spines and stiff bristles, but the development of stout spines is not nearly so pronounced as in spivalicovnis , and true spines do not occur on the outer surface of the chelicera. Text fig. 3. Solpuga stvepsicevos Kraepelin. Shewing terminal portion of right upper jaw, with flagellum: specimen from Barberton. The splendid colouration of this species — head-plate, palps, legs and sides of abdomen clothed with bright, golden yellow hairs, tergites black — serves to distinguish the female from those of related species. The margins of the malleoli are broadly infuscated but not deeply so. The measurements of the female are as follows: breadth of head-plate 10*7, length of tibia and tarsus of palp 10, of patella of fourth leg 9, of tibia of fourth leg 7-8. Solpuga schonlandi Pocock [Text-fig. 4], 1900. Ann. Mag. Nat. Hist. 7, vi. p. 316. The type was recorded from Grahamstown, but there are no specimens in the Albany Museum from the Grahamstown neighbourhood. I have previously recorded it from Kimberley, from the neighbourhood of Johannesburg, and from several localities in the Pretoria district. It is also known to me from Rooispruit near Rosmead (A. Gib- bons) ; Tafelberg (Miss A. Gadd) ; Damplaats near Burghersdorp (A. Kruger) ; Vryburg, Warrendaleand Kuruman (Kimberley Mus.) ; Modder Riv. (F. A. O. Pym) and Bulawayo (Miss L. Leppan). In the male, there may be either five or four teeth in the single series of the upper jaw: in the former case the third tooth is minute. The terminal fang is rather long, curved slightly outwards towards the apex, and a little downwards at the apex : on the inner edge superiorly is a small sharp-pointed forwardly projecting tooth, situated much nearer to the flagellum than to Text fig. 4. Solpuga schonlandi Pocock. Portion of left chelicera of male, viewed from mesial side : specimen from Rooispruit. 30 Annals of the Transvaal Museum the apex of the fang. The lower jaw has no distinct lateral keel on its outer side distally: the distance from the tip of the fang to the apex of the first tooth is about equal to if— if times the distance between the apices of the two large teeth. The basal enlargement of the flagellum is high, not produced anteriorly: the upper margin is acute, but does not form a sharp keel well marked off from the outer turgid portion: on its outer side it is flanked by a strong chitinous thickening which is continued obliquely downwards into the fang. The shaft is short, lightly curved, considerably flattened at the anterior bend in an antero-posterior direction, but elsewhere is more or less cylindrical, gradually diminishing in width towards the apex, which is truncate with a central projecting core. In dried specimens the shaft bends downwards over the basal enlargement, and thus the apex becomes carried far back, distinctly behind the hind margin of the basal enlargement. There is a fairly strong development of spines and stiff bristles on the upper and outer surfaces of the chelicera. Colour: head-plate and appendages pale yellowish, the front margin of the former infuscate : abdominal tergites pale brown : malleoli not infuscated. The female has four or five teeth in the single series of the upper jaw: upper and lateral surfaces of the chelicera more or less distinctly marked with three darkish longitudinal stripes: hairs of abdomen greyish. This latter character will separate it from the female of Venator in which the hairs of the abdomen are uniformly lemon yellow. So far as I can discover, the female presents no structural or colour characters which will serve to distinguish it from globicornis which is common in the Pretoria district, or from ferox which occurs in the Free State and Transvaal. However, the relation between the width of the head-plate and the length of the several segments of the fourth leg may perhaps ultimately furnish a guide to the specific identity of adult examples. Measurements. Total length, M, 41: width of head-plate, F, 12-85: length of tibia and tarsus of palp, M, 13-5, F, 11-85 : of patella of palp, M, 13, F, io-8: of tibia of fourth leg, M, io-8, F, 9: of patella of fourth leg, M, 11-2, F, 9-6. Solpuga globicornis Kraepelin, 1899 [PI. V, fig. 21]. Das Tierreich, p. 76, fig. 47. This species has been previously recorded by me from various localities in the Pretoria district, and with some doubt from the Lydenburg and Zout- pansberg districts. In the lower jaw of the male the distance from the tip of the fang to the apex of the first tooth is equal to about if times the distance between the apices of the two large teeth. In the female, the patella of the fourth leg seems to be always longer than the tibia. Measurements of several females apparently referable to this species are as follows: Width of Length of tarsus + Length of Length of Length head- tibia of patella patella of tibia plate palp of palp IV IV Immature example from Pretoria 10-9 n-3 10-4 9-5 8-8 Adult from Gezina I2-I 12 11-25 10 9-4 Adult from Wonderboom ... 12-5 ii-8 10-65 9-7 9-3 Only very few cylinder bristles occur on the tibia of the female palp: there are some also on the tarsus. Annals of the Transvaal Museum 3i Solpuga sericea Pocock [PL III, fig. 6], 1897. Annals Mag. Nat. Hist. 6, xx. p. 260, fig. 4. The types came from Gadzima on the Umfnli River, Mashonaland. Mr S. Hirst records the species from Petauke and from Alala Plateau, localities north of the Zambesi. It is recorded by Dr Purcell from Shilowane. We have a female example from Salisbury (Fr. J. O’Neil, S.J.) which is probably referable to this species (or possibly to celeripes Hirst). There are numerous very long silky hairs on the hind legs, a very unusual character in females: the hairiness of the hind legs is indeed more pronounced than in females of chelicornis. The hind borders of the genital sternites are broadly rounded. There are two intermediate teeth in the single series of the upper jaw. Measurements: breadth of head-plate 5-1, length of tibia and tarsus of palp 6*i, of patella of palp 5*5, of patella of fourth leg 5-7, of tibia of fourth leg 57- Solpuga celeripes Hirst [PI. VI, fig. 32], 1911. Manchester Memoirs, lvi. No. 2, p. 10, fig. 2. The type came from Salisbury, S. Rhodesia. The adult male is the smallest known to me in this genus, the type being 11 mm. long and its head-plate 3 mm. broad. A specimen in the Transvaal Museum has the following measure- ments: breadth of head-plate 3*1, length of tibia and tarsus of palp 5-7, of patella of palp 5-4, of tibia of fourth leg 5-7, of patella of fourth leg 5-3. In describing the species, Mr Hirst remarked that the dark markings and general colouration are almost exactly the same as in 5. sericea, but the narrow yellow stripe which is present on either side of the dark central band of the dorsal surface of the abdomen in sericea seems to be absent (the abdomen being shrunken) . In the specimens examined by me, however, the continuous yellow stripes are clearly present. The adult male is remarkable for the relative shortness of the fang of the lower j aw, the distance from the tip of the fang to the apex of the first tooth being subequal to the distance between the apices of the two teeth. Solpuga alstoni Purcell, 1901. Annals S. Af. Mus. 11. p. 209, fig. 1. The type was taken at Eities in Gt. Bushmanland. Solpuga darlingi Pocock, 1897. Ann. Mag. Nat. Hist. 6, xx. p. 259 and fig. 5, p. 261. The type came from Gadzima on the Umfuli River, Mashonaland. Solpuga ferox Pocock [Text fig. 5], 1895. Ann. Mag. Nat. Hist. 6, xvi. p. 83, PI. 4, fig. 3. (Kraepelin’s figure in Das Tierreich (p. 71) represents the characters of the male chelicera more correctly than that accompanying Pocock’s original description, but is also somewhat misleading.) The type is labelled Port Elizabeth, but the record requires confirmation in my opinion. The species has been recorded by me from Kimberley and Rustenburg, and I know of it also from Venterskroon (M. H. Viljoen) ; Kroonstad (Miss D. Chennells); and Bloemfontein (Dr T. F. Dreyer). The Kimberley Museum has it from Fourteen Streams, Barkly West, Taungs, Pniel, and Wirsing Siding as well as from Kimberley. The male has a distinct keel on the outer side of the lower jaw distally. Solpuga sagittaria Pocock, 1900. Ann. Mag. Nat. Hist. 7, v. p. 299, figs. 5 and 5 a. See also Das Tierreich, p. 74, figs. 42 and 43. The locality cited for the type is Mazoe, Mashonaland. 32 Annals of the Transvaal Museum Solpuga suffused Hewitt [PI. V, fig. 22], 1916. Annals Durban Mus. 1. p. 217. The locality of the type is unknown. In both upper and lower jaw there is an exceptionally strong development of feather bristles. The fang of the lower jaw is short, the distance from the tip to the apex of the first tooth being about equal to i| times the distance between the apices of the two large teeth. Solpuga cervina Purcell, 1899. Annals S. Af. Mus. 1. p. 415, figs. 21 and 21 a, also 11. p. 208. The type came from Clan william: other specimens were recorded by Purcell from Stein kopf in Namaqualand, and females apparently referable to this species from Van Wyk’s Vlei and from Namies in Bushmanland. Text fig. 5. Solpuga ferox Pocock. Shewing flagellum and terminal portion of left upper jaw viewed (a) from the mesial side, (&) from the dorsal side: specimen from Kimberley. Solpuga alcicornis Kraepelin [PI. VII, fig. 34], 1914. Beit. z. Kennt. Land- u. Susswasser fauna Deutsch-Sudwestafrikas , Skorpiones u. Solifugae, p. 125, fig. 2. The type came from Keetmanshoop. It is also known to me from the neighbourhood of Kuruman (F. A. O. Pym) and Mt Temple (T. C. Lanham). In the Kuruman specimens, the flagellum, and with it the terminal fang of the upper jaw, is twisted outwards away from the main axis of the jaw. The basal enlargement is high and swollen. There are long stridulatory ridges on the chelicerae. The upper surfaces of the chelicerae bear long stout bristles but no definite spines. The fang of the lower jaw is short, the distance from the tip to the apex of the first tooth being about 1^ times as long as the distance between the apices of the two large teeth. Annals of the Transvaal Museum 33 Solpuga collinita Purcell, 1903. Annals S . Af. Mus. hi. p. 3, fig. 2. The type came from Willowmore C. P. Solpuga vincta C. L. Koch, 1842. Kraepelin, Das Tierreich, p. 63, fig. 23. Purcell, Annals S. Af. Mus. 1. p. 420, fig. 23. It is recorded by Dr Purcell only from the neighbourhood of Capetown. Solpuga maraisi Hewitt [PI. V, fig. 23], 1913. Records Albany Mus. n. p. 480, Text fig. The type came from Caledon C. P. The species is also known from Worcester (G. B. Townshend), and Stellenbosch (C. S. Grobbelaar). On the outer side of the lower jaw, a keel is present near the apex of the fang, but more proxi- mally the lateral crest is obsolete. The feather bristles of the upper and lower jaws are weakly developed. The fang of %the lower jaw is not greatly elongated, the distance from the tip of the fang to the apex of the first tooth being about twice, or a little less than twice, the distance between the apices of the two large teeth. Stridulatory ridges are moderately well developed on the upper jaw: Solpuga spectralis Purcell, 1899. Annals S. Af. Mus. 1. p. 424, fig. 25. The type specimen came from Klipfontein, Namaqualand division. Solpuga lateralis C. L. Koch [PI. V, fig* 24], 1842. Kraepelin, Das Tierreich, p. 61, fig. 19. Purcell, Annals S. Af. Mus. 1. p. 425, fig. 26. Dr Purcell records this species from Port Elizabeth. It is known to me fromDunbrody (Fr. Vogt), and from Bussacks near the Kareiga River mouth (Mr F. G. C. Graham) . In males of this species, the stridulatory ridges of the upper jaw are rather weakly developed, sometimes being nearly obsolete. The fang of the lower jaw is moderately elongated, the distance from its tip to the apex of the first tooth being equal to about 2 1 times the distance between the apices of the two large teeth. The fang of the upper jaw is without a keel on its mesial side, as occurs in hostilis and derbiana, but the keel is represented by a prominent Text fig. 6. Solpuga erythronotoides sp. nov. Den- sharply pointed tooth. tition and flagellum of male, viewed from the mesial side. Solpuga erythronota Kraepelin, 1900.. Das Tierreich, p. 64, fig. 24. The type of the species is indefinitely located S. Africa. Solpuga erythronotoides sp. nov. [Text fig. 6]. This species is founded on a single male example found at Victoria West by Mr B. Marais. It is very closely related to 5. erythronota Kraepelin, but 3 34 Annals of the Transvaal Museum seems to differ as follows : the flagellum is shorter and its anterior bend more forwardly situated : there is a distinct interval between the first and second teeth of the lower jaw. Dentition . The terminal fang of the upper jaw is short: there are three distinct anterior teeth, the first being small and the second and third of moderate size : there follows a long toothless interval terminated by the fourth tooth which is small and the fifth which is large but not much larger than the third : in the double series, the outer row has four teeth but the inner one only three, of which the distal one is largest. On the inner side of the jaw dorsally, near to the terminal fang and just in front of the anterior bend of the flagellum, is a prominent sharp pointed tooth terminating an abbreviated keel or ridge. The lower j aw has three teeth, the distal one largest and longest and separated from the small intermediate tooth by a short interval. On its outer side, a sharp lateral keel is only present in the distal fourth, whence it is continued as a line of granules to the base of the jaw. The distance between the tip of the fang and the apex of the first tooth is equal to about 2\ times the distance between the apices of the two large teeth. Flagellum. The basal enlargement is well elevated and has a sharply keeled dorsal margin: anteriorly it is only a little produced, the anterior bend being situated above the third tooth: the recurrent portion is sub- cylindrical and slender, except at the anterior bend where it is strongly flattened from front to back though not quite so broad as the fang at this point: it passes in a light curve immediately above the basal enlargement, and, twisting slightly outwards, terminates in an acutely pointed apex a little posterior to the hind margin of the basal enlargement; the length of the flagellum occupying a post-laminar position being much less than the distance between the anterior bend and the hind margin of the lamina. The stridulatory area of the chelicera is well developed. Colour. Head-plate and appendages pale brown: tibia and more distal segments of fourth leg dark brown, and the distal segments of the other legs and of the palp are also more darkly coloured than the basal segments: abdominal tergites pale brown with some infuscation laterally, but there is no strongly contrasting pigmentation on the tergites : sides of abdomen clothed with long pale hairs: malleoli broadly margined with black. Measurements. Total length 23, length of flagellum 1*7, of patella of palp, 5-4, of tarsus and tibia of palp 6-2, of patella of fourth leg 5-75, of tibia of fourth leg 5*2. This species, and its near ally erythronota, are nearly related to 5. lateralis. Solpuga derbiana Pocock [PI. II, fig. 4 and Text fig. 7], 1895. Ann. Mag. Nat. Hist. 6, xvi. p. 90, PI. IV, fig. 8. The type is labelled “Interior of S. Africa. ” Dr Purcell has identified therewith a species common at Grahamstown, and the information kindly supplied to me by Mr S. Hirst regarding the characters of the type specimen seems to confirm the identification. The species is known to me from Grahams- town, Brakkloof (Mrs G. White), Peddie and Line Drift near Peddie (B. Marais), Majuba Nek, Herschel dist. (J. Hepburn), and Damplaats near Burghersdorp (A. Kruger). The length of the recurrent portion of the flagellum is greater than twice the distance from the anterior bend to the hind margin of the basal enlarge- ment: and the tip of the flagellum reaches to a point situated very much Annals of the Transvaal Museum 35 nearer to the hind dorsal margin of the chelicera than to the hind margin of the basal enlargement. The length of the flagellum will probably prove to be somewhat variable, but I have not yet seen material that can be regarded as intermediate between this species and coquinae, which is mainly distin- guished therefrom by its shorter flagellum. The basal enlargement of the flagellum is considerably elongated, and the anterior bend, which is not markedly broadened, lies immediately over the second tooth, or even very slightly anterior thereto. The shaft is subcylindrical almost throughout, but towards the tip there suddenly appears a dark brown sharp cutting edge dorsally: this is usually very slightly raised above the rounded surface of the main portion of the shaft, but to a variable extent, the cutting edge being more prominent in Peddie specimens than in examples from Grahams town, and is scarcely visible in a specimen from Brakkloof near Grahamstown. The terminal fang of the lower jaw is not long, the distance from its tip to the apex of the first tooth being equal to about if-if times the distance between the apices of the two large teeth. In the female the anterior tergites are reddish brown, becoming dark brown or nearly black near their lateral borders: in the sixth tergite the Text fig. 7. Solpuga devbiana Pocock. Shewing flagellum and terminal portion of left upper jaw viewed from mesial side: specimen from Peddie. posterior border is also infuscated, and succeeding tergites are quite black. Sides of abdomen with pale hairs. Sternites dark brown at their lateral margins, but only very narrowly so anteriorly, and more broadly so in the posterior sternites. Hind legs red, chelicerae reddish, and head -plate with a red tinge. In the male the general colour is dull drown and the tergites are very dark, black behind and at the sides: mesial portions of six anterior tergites brown. Hind limbs brown, with well developed mane, the hairs thereof white with j ust a tinge of pale violet. Scopula of palp rufous. Sides of abdomen whitish. A subadult male, lacking the flagellum, taken in Grahamstown (15. xii. 1918), greatly resembles the female in colour: it differs therefrom in the well developed mane of the hind leg, and the sides of the abdomen are whiter than in the female. Measurements of adult male from Peddie and of adult female from Grahams- town: breadth of head-plate, M, 7-1, F, 8-6: length of patella of palp, M, 7-7, F, 6-7: of tibia and tarsus of palp, M, 7 -8, F, 7-6: of patella of fourth leg, M, 6-9, F, 6-8: of tibia of fourth leg, M, 6-25, F, 6-1. Solpuga tookei sp. nov. [Text fig. 8]. The type of this species is a single adult male from Mariannhill, Natal, kindly presented to the Albany Museum by the Curator of the Museum at the 3—2 36 Annals of the Transvaal Museum Trappists Monastery near Pinetown. It is named after Lt. W. M. B. Tooke to whom the Albany Museum is indebted for the identification and arrange- ment of the collection of ticks belonging to that institution1. The species belongs to the group including hostilis White, and marshalli Poc., both of which have been recorded from the Durban neighbourhood (but the latter species quite erroneously, I think). In the Mariannhill specimen, the second tooth of the upper jaw is quite large, and the distance from the tip of the fang to the apex of this tooth is subequal to the distance between the apices of the second and fourth teeth: the gap between the second and third teeth is in fact not nearly so long as in 5. marshalli Pocock or 5. hostilis White, which latter species it more closely resembles in the characters of the flagellum. Dentition. Upper jaw with a terminal fang of moderate length, and not upturned: first tooth small, second large, then follows a rather short and shallow gap, third tooth rather small, fourth the largest: in the double series, the outer row comprises three moderate sized teeth and one small one basally situated, whilst the inner row has the first and third teeth of moderate size Text fig. 8. Solpuga tookei sp. nov. Portion of left upper jaw, with flagellum, viewed from mesial side. but the second and fourth minute. In the lower jaw there is a strong prominent crest on the outer side, extending from apex to base. The two major teeth are both large and the middle one rather small : between the large distal tooth and the small middle one, there is a short but well-defined interval. The fang of the lower jaw is not long, the distance from the tip to the apex of the first tooth being about if times as long as the distance between the apices of the two large teeth. The stridulatory area on the inner surface of the upper jaw is ill developed, the ridges being short and weak, and the whole area decidedly smaller than usual: there are five ridges present and rudiments of two others. Flagellum. The basal enlargement is moderately elongated; the anterior bend is in the same vertical as the first tooth; the shaft is narrow and sub- cylindrical, passing backwards just above the basal enlargement and extending to a point which is a trifle more remote from the hind margin of the basal enlargement than this is from the tip of the fang, thus not reaching so far as the hind margin of the chelicera. At the anterior bend the flagellum is not broadened, its width being less than half the extreme width of the fang at this point. Near the tip of the flagellum it presents dorsally a sharp cutting edge for a short distance. 1 Since this was written, my friend William M. B. Tooke, 2nd Lieut. South African Infantry, fell in action at Fampoux, 12th April, 1917. Annals of the Transvaal Museum 37 Palp with the tibia scopulate below except near the base, and only thinly scopulate near the apex. Posterior legs carrying long hairs but they do not seem to form a definite mane. Colour. Head-plate, mandibles, and appendages pale brown, the more distal segments of palps and legs dark brown, the infuscation being strongest on the distal segments from the patella onwards of the fourth leg. Abdominal tergites very dark, almost black superiorly throughout, but sides of abdomen clothed with pale yellow hairs. Malleoli not infuscated. Measurements. Total length 20, length of flagellum 5-2, of patella of palp 5-25, of tarsus and tibia of palp 6, of patella of fourth leg 4-7, of tibia of fourth leg 4*55- Solpuga coquinae Hewitt, 1914. Records Albany Museum, hi. p. 9, fig. 2. The type came from Cookhouse C.P. and we have other specimens from Somerset East (E. Driver), Longhope (E. Abrahamson) and a fairly distinct variety from Kimberley (J. H. Power) and Bloemfontein (Dr T. F. Dreyer). The flagellum varies a little in length, being a trifle longer relatively in large specimens than in small ones, but the total length of the recurrent portion is never more than twice the distance from the anterior bend to the hind margin of the basal enlargement: the apex is considerably nearer to the hind margin of the basal enlargement than to the posterior dorsal margin of the chelicera. The flagellum is not markedly broadened at the anterior bend. In specimens from Kimberley and Bloemfontein the flagellum differs only slightly from that of the type, in that the basal enlargement is a trifle deeper, more rounded, and less produced anteriorly, whilst the shaft is a little longer and straighter but sometimes bent downwards rather abruptly near the tip: these minor differences are exhibited more particularly in the single example from Bloemfontein. It is possible, however, to distinguish this form from that of the type, owing to the greater length of the fang of the lower jaw. In the typical variety, the distance from the tip of the fang to the apex of the first tooth is about equal to i|— if times the distance between the apices of the two large teeth: in the Kimberley and Bloemfontein specimens, the proportion is greater, being 2-2 1 : 1. This form I now designate 5. coquinae var. nov. orangicus. The species is smaller than 5. hostilis or derbiana : the measurements of an adult female from Cookhouse are as follows: breadth of head-plate 6-75, length of tibia and tarsus of palp 6-7, of patella of palp 5-6, of tibia of fourth leg 5*6, of patella of fourth leg 6. The colouration is very much like that of hostilis. In the typical male all the abdominal tergites are infuscated throughout, and the sides of the abdomen silvery, but in the Kimberley and Bloemfontein specimens, one or two of the anterior tergites are dark brown not black. The hind legs are more or less infuscated, rather than red as in hostilis. In the female, on each side of the anterior tergites, which are brown, there is a more or less distinct dark stripe passing backwards into the blackened area posteriorly, and along the midline also is a series of dark stripes but these are in the soft skin between the tergites : the hind legs are red. 38 Annals of the Transvaal Museum Solpuga hostilis White [PL IV, figs. 16 and 17, and VI, fig. 29], 1846. Pocock in Ann. Mag. Nat. Hist. 6, xvi. p. 89, PI. IV, fig. 7. Purcell in Annals S. A f. Mus. 1. p. 427, fig. 27 ( cultvata ). The species cannot be recognised from the original description and figure given in Methuen’s Life in the wilderness, and the locality of the type was simply stated as “S. Africa, near the tropic of Capricorn.” Mr Pocock, having examined the types, has specifically identified therewith some specimens from Estcourt; and a species recorded from Durban and described by Dr Purcell under the name of 5. cultvata is now regarded by Purcell and Kraepelin as hostilis. This species is common in the Transvaal and has been recorded by me from various localities in the Pretoria and Zoutpansberg districts. We have a series of both sexes from Doornkop near Belfast (R. Gerhardt). An example from Liineburg, Natal (W. Oom) differs from any of the above in that the tip of the upper fang is scarcely up-turned, and the flagellum is not bent downwards near the apex. In the male, stridulatory ridges are usually quite absent from the chelicerae : in a specimen from Gezina, Pretoria, three abbreviated ridges occur. The fang of the lower jaw is long, the distance from its tip to the apex of the first tooth being equal to about 2^ times the distance between the apices of the two large teeth. The keel on the outer side of the lower jaw of the male is very strongly developed, being much more pronounced than in. marshalli. It varies however to some extent, but usually in the basal portion of the jaw is raised into quite a high ridge. In the female the anterior tergites are uniformly reddish brown, the blackening of the posterior part of the abdomen commencing rather suddenly: in Transvaal specimens black pigmentation only occurs on the last three tergites. In specimens from Liineburg the sides of the anterior tergites are infuscated, but this blackening does not take the form of sharply defined blotches or stripes such as occurs in chelicornis. Males have the tergites more or less infuscated throughout, the mesial portions of the anterior tergites being dark brown and the sides black: hind legs red in both sexes, maned only in the male. The measurements of an adult female from Pretoria are: breadth of head-plate 9-1, length of tibia and tarsus of palp 8, of patella of palp 7, of tibia of fourth leg 6-5, of patella of fourth leg 7-3. A very young specimen from Mfongosi, Zululand, has only 5 joints on the fourth tarsus of one side, but is normal on the other side. Solpuga marshalli Pocock [PI. VI, fig. 30], 1895. Ann. Mag. Nat. Hist. 6, xvi. p. 91, PI. IV, figs'. 9 and 9 a. The type came from “Fort Salisbury, Mashonaland.” The species is not easily recognised from Pocock’s figures, whilst Kraepelin’s illustration in Das Tierreich seems to me referable to some other species, and perhaps may not be based on Mashonaland material. Kraepelin records the species from the Transvaal and from Durban but these records may be viewed with some suspicion in view of possible confusion with hostilis, and seeing that no specimens of marshalli are represented in the Transvaal and Durban Museums. Mr Hirst informs me that in the type specimen the end of the flagellum is very finely pointed, and the edges near the end very finely granular. We have male and female specimens from Salisbury (Fr. J. O’Neil, and C. von Hirschberg). In the male, as in the female, stridulatory lamellae are present on the chelicerae but are not long. Towards the tip of the flagellum its dorsal edge is minutely serrulated. The fang of the lower jaw is not very Annals of the Transvaal Museum 39 long in the male, the distance from the tip of the fang to the apex of the first tooth being about equal to if times the distance between the apices of the two large teeth. The measurements of male and female specimens are as follows: breadth of head-plate, M, 7-5, F, 9: length of tibia and tarsus of palp, M, 8-5, F, 7-9: length of patella of palp, M, 8-35, F, 6-65 : length of tibia of fourth leg, M, 7-8, F, 6-3: length of patella of fourth leg, M, 8-4, F, 7-1. This species is more darkly coloured than hostilis. The tergites are strongly infuscated in both sexes, but the mesial portions of the anterior segments are dark brown rather than black. The soft skin at the sides of the abdomen is somewhat infuscated, though silvery white hairs occur there. The sternites are broadly infuscated at the sides in both sexes, but more specially in the male. Hind legs very dark brown. Soipuga junodi Purcell, 1903. Novitales Zoologicae, vol. x. p. 304, fig. 2. The type came from Shilowane, Zoutpansberg dist. and I have recorded it from Vygeboompoort in the Waterberg dist. Soipuga tubicen Kraepelin [PL VII, fig. 35], 191 1 . Mit. a. d. Nat. Mus. Hamburg, xxviii. p. 102. This is a Transvaal species, but no precise locality is known. There is a dense group of stout bristles on the upper surface of the chelicera, near to the basal enlargement, and to the end of the shaft. On the inner side of the upper fang there is a very strong outstanding keel, commencing at the anterior bend and extending forwards beyond the first tooth. No distinct mane on the hind legs. In the lower jaw, the distance from the tip of the fang to the apex of the first tooth is equal to about twice the distance between the apices of the two large teeth, or a trifle more than twice. Soipuga hamata Hewitt, 1913. Annals Transvaal Mus. iv. p. 160, fig. 22. The type came from Mamiaanshoek near Zwagershoek in the Waterberg dist., and other specimens were recorded from Vygeboompoort in the same district. The fang of the lower jaw is very long, the distance from the tip to the apex of the first tooth being about three times the distance between the apices of the two large teeth, or even a little more. Soipuga bechuanica Hewitt [PI. IV, fig. 14], 1913. Annals Transvaal Mus. iv. p. 1 61, fig. 23. The type came from Serowe in the Bechuan aland protectorate. Stridula- tory ridges on the chelicerae are only weakly developed in the male. The large teeth of the lower jaw are widely separated, a distinct interval occurring between the distal tooth and the small intermediate tooth: the distance from the tip of the fang to the apex of the distal tooth is about if times the distance between the apices of the two large teeth. As in chelicornis and villosa, this species has a row of four long spines on the outer side of tarsus II superiorly: in hostilis and Venator distinct spines are wanting, being represented however by bristles several of which in marshalli are spiniform. Soipuga chelicornis A. Licht. [PI. Ill, fig. 11], 1796. Kraepelin, Das Tierreich, p. 59, fig. 17. Of this species, several varieties markedly differing in colour may be recognised. 40 Annals of the Transvaal Museum Mr S. C. Cronwright-Schreiner, in writing of the living animal as found at Hanover, described it as "of a most brilliant yellow, with a heavy black band down the back of the abdomen while the legs are covered with long yellow hair, which in the male, becomes a distinct mane and is iridescent. As it lies on the sand on a hot day, sparkling in the sunshine, it is a most exquisite creature1.” This variety I now designate 5. chelicornis var. pubescens [PI. II, fig. 2]. Its most marked character is a dense pile of short pale hairs on the upper surfaces of the patella and tibia of the palp in the adult male, and to a less extent in the female. The sides of the head-plate and chelicerae are thickly, though rather shortly, bearded. The fringes of hair on the hind legs are particularly heavy, extending in attenuated form as far as the distal segments of the tarsi. The lateral black bands of the abdomen commence to merge on the fourth abdominal tergite, and fusion is quite complete on the fifth tergite. The terminal fang of the lower jaw is long, the distance from the tip of the fang to the apex of the first tooth being equal to about twice the distance between the apices of the two large teeth. In the single series of the upper jaw, the distance between the first and second teeth is about 1^ times the distance between the second and fourth teeth. This variety is known to me from De Aar (S. C. Cronwright-Schreiner and Miss E. Friedlander), and presumably the Hanover specimens are referable thereto. We have female examples which seem to belong to the same variety from Dikkop Flats near Grahamstown (Miss J. Harris), and from Zandkraal near Steytlerville (Miss A. Geard). Also, Dr Purcell’s records from Namaqualand, Kenhardt, and Willowmore divisions are probably based on this variety. According to Kraepelin’s account in Das Tierreich, the typical form of the species is coloured quite differently, much as in the following form now named 5. chelicornis var. rufescens [PI. II, fig. 3]. This variety is known to me from Longhope (Miss E. Abrahamson), and Dirkskraal, Somerset East dist. (B. Marais). The upper surfaces of the patella and tibia of the palp are quite devoid of the pile of short hairs that occurs in pubescens. The sides of the head -plate and chelicerae are not bearded. All the hairs of the mane on the hind legs are. white, at any rate in their distal portions, but the hairs situated on the broader part of the mane are magenta coloured in the basal half of each hair. The hairs on the sides of the abdomen are quite white. The surfaces generally are dull brown, more darkly so on the tibia and tarsus of the palp: the pale brown mesial area of the abdomen superiorly extends over the first seven tergites, being bordered on each side by a black stripe, and posteriorly by a black patch which covers the hind tergites. The jaw characters are similar to those of pubescens. Another distinct variety occurs at Kakamas (Miss H. C. Olivier). It differs from the two just described principally in the spacing of the teeth of the single series of the upper jaw: the distance between the first and second teeth is less than the distance between the second and fourth teeth; in the lower jaw, the distance from the tip of the fang to the apex of the first tooth is about equal to if times the distance between the apices of the two large teeth. The appen- dages are pale yellowish (in spirits) : the posterior tergites are not so deeply blackened as in pubescens, yet the infuscation extends considerably forwards, the mesial pale brown area only reaching backwards over the first four abdominal tergites. The mane of the hind legs is not so strongly developed 1 "Some Arachnids at Hanover, Cape Colony,” by S. C. Cronwright-Schreiner, in the Popular Science Monthly, December, 1902. Annals of the Transvaal Museum 4i as in either of the varieties just described, nor is there a conspicuous pile of short hairs on the upper surfaces of the patella and tibia of the palp, although numerous very short hairs occur there. This form I now designate 5. chelicornis var. nov. macrognathus [Text fig. 9]. The colouration approaches that of the hosiilis section, and the dentition is not very different from that of 5. beckuanica. Lastly, Kraepelin has recorded chelicornis on the evidence of a female specimen from Okawango in the north of S.W. Africa. It may also be noted that Kraepelin’s figure in Das Tierreich seems to indicate a variety distinct from any of the above in respect to the dentition. Text fig. 9. Solpuga chelicornis macrognathus var. nov. Left chelicera of male viewed from mesial side. The measurements of the adult males of the three forms here distinguished are as follows : var. rufescens var. pubescens var. macro- gnathus Breadth of head -plate 6-i 8 5-8 Length of patella of palp 8-2 io-8 8-3 Length of tibia and tarsus of palp 8-8 ii-i (absent) Length of patella of fourth leg . . . 8-3 ii-i 8-5 Length of tibia of fourth leg 8-4 I1'3 8-15 Solpuga villosa Purcell, 1899. Annals S. Af. Mus. 1. p. 422, fig. 24. The type male of this species was indefinitely located “S. Africa.” Other specimens which seemed to be referable thereto were recorded from Concordia in Namaqualand, and from Fraserburg: these, however, were apparently female examples, and as such possibly indistinguishable from chelicornis. Subsequently, Dr Purcell regarded this species as essentially the same as chelicornis, and Prof. Kraepelin al%o held the two as cospecific, though distinct as varieties. To me, it seems well worthy of distinction from chelicornis. We have male specimens agreeing precisely with the description of villosa, from Victoria West (P. D. Morris), and from Tafelberg (Miss A. Gadd). Females from these localities are easily distinguishable from those taken along with the pubescens form of chelicornis in the colour pattern of the 42 Annals of the Transvaal Museum abdomen : in villosa, there are only three pairs of black blotches on the abdomen anteriorly, these being followed immediately by the black patch which covers all the posterior tergites: in chelicornis var. pubescens there are six or seven pairs of black blotches in front of the continuous bands. In both species, the posterior abdominal sternites are infuscated throughout their breadth, instead of at the sides only, as in females of hostilis or derbiana : the blackening is more intense at the sides however. The hind borders of the genital plates are considerably produced in both species. The terminal fang of the lower jaw is not quite so elongated as in the pubescens form of chelicornis , which this species resembles in the possession of a dense pile of short hairs on the upper surfaces of the distal segments of the palp in both sexes, but more especially in males: the distance from the tip of the fang to the apex of the first tooth is about equal to if times the distance between the apices of the two large teeth. Solpuga hastata Kraepelin, 1899. Das Tierreich, p. 58, figs. 15 and 16. The type and only known specimen is indefinitely located Gt. Namaland. Solpuga methueni Hewitt, 1913. Annals Transvaal Mus. vol. iv. p. 153, fig. 18. The type came from Quibis, S.W.A., near the Karasbergen. Solpuga ornithorhyncha Hewitt [PL V, fig. 19], 1913. Annals Transvaal Mus. vol. iv. p. 151, fig. 17. The types came from localities near the Karasbergen, S.W.A., viz. Kraikluft, Narudas Sud, and from between Kraikluft and Sandmund. Stridu- latory ridges are well developed on the chelicerae of the male. The lower jaw has numerous feathered bristles on its inner side, and on the outer side is a feeble but distinct ridge distally. The colouration of the abdomen has a general resemblance to that of hostilis, and the species is no doubt diurnal in habit. The tergites are entirely brown except for slight infuscation in the middle, which is faint in the anterior segments but more pronounced posteriorly, the three posterior tergites being dark brown throughout: the soft skin between the tergites is also blackened mesially. On each side of the tergites, the soft skin is blackened as a continuous longitudinal stripe. Below this, the sides are silvery. The sternites are infuscated laterally. Solpuga lineata C. L. Koch [Pis. Ill, fig. 7, and VI. fig. 33], 1842. Kraepelin, Das Tierreich, p. 65, fig. 26. Purcell, Annals S. Af. Mus. 1. p. 428, fig. 28. Dr Purcell records this species from the divisions of Namaqualand, Carnarvon, Uitenhage, Robertson and Swellendam. It is known to me from Alicedale (F. Cruden), Dunbrody (V. Powels), De Aar (Miss E. Friedlander) and Victoria West (B. Marais). Stridulatory ridges are well developed on the chelicerae of the male. In the lower jaw of the male, the patch of bristles on the inner surface includes two or three curved spines distally, thus differing from hostilis, and most other species, where the patch is composed of bristles of more uniform size : the stouter curved bristles and spines are mostly in a single line along the lower portion of the extensive patch of bristles. There are numerous feather bristles which, however, like those of the upper jaw, are not very densely feathered. There is a very long interval between the first and second teeth of the lower jaw. On the outer side of the lower jaw is a distinct keel. In the female the posterior margins of the genital sternite are rounded. Annals of the Transvaal Museum 43 Species incertae sedis. Solpuga brevipalpis Purcell, 1899. Annals S. Af. Mus. 1. p. 431, fig. 29. The types are female specimens from Naroep in Gt Bushmanland. It is an ally of 5. lineata. Solpuga cajfr a Vocock, 1897. Ann. Mag. Nat. Hist. 6, xx. p. 262. The types, two adult females, came from Estcourt. The colour characters are like those of S. toppini Hirst, from Ngxwala Hill, Zululand. Solpuga nigrescens Pocock, 1895. Ann. Mag. Nat. Hist. 6, xvi. p. 88. The type is a female example labelled ‘ ‘ Lower Zambesi. ' ’ Mr Pocock doubt- fully identified therewith some specimens from the north-eastern region of Victoria Nyanza. Solpuga schultzei Kraepelin, 1908. Denks. d. med.-nat. Gesell. Jena, xm. p. 270, figs. 2 and 3. This species is based on a female example collected at Rooibank near Walfish Bay. It is closely related to picta — which according to Kraepelin includes nigrobraccata — the two species being remarkable in the great elonga- tion of the jaws, the first tooth of the upper jaw being considerably removed from the second as well as from the apex of the jaw. Apparently also, the tibia of the second leg has thickened hairs instead of spines on the dorsal side externally. Solpuga striata Kraepelin, 1914. Beit. z. Kennt. Land- u. Susswasserfauna Deutsch-Sudwestafnkas, Skorpiones u. Solifugae, p. 124, fig. 1. The type specimen, taken on farm Voigtsland about 38 km. east of Windhuk, is probably very immature, being only 11 mm. long (without mandibles). Kraepelin seemed to regard it as referable to the group of 5. lateralis and allies, but I have no doubt that it really belongs to the group of species including 5. sericea Poc., 5. zebrina Poc. and 5. celeripes Hirst, and quite possibly is identical with one of these. The colour pattern represented in Kraepelin’s illustration is precisely similar to that exhibited by male and immature specimens of celeripes. Solpuga picta Kraepelin, 1899. Das Tierreich, p. 81, figs. 59 and 60. Purcell, Annals S. Af. Mus. 1. p. 431, fig. 30 (5. nigrobraccata) . The species is merely located as Damaraland. Solpuga scopulata Karsch, 1880. Kraepelin, Das Tierreich, p. 60. This species is only known from a female specimen taken at Han tarn C.P. : it seems to be very like chelicornis Licht. Key to the South African species of the genus Solpuga Licht. mainly based on the characters of adult males. Group I. Dentition of upper jaw almost alike in the two sexes, the distal series composed of four or five teeth in a continuous row, not broken by long toothless intervals, the first and second teeth being relatively large. The 44 Annals of the Transvaal Museum terminal fang of the lower jaw not keeled on the outer side, or only weakly so, and not greatly elongated in the adult male, the distance from the tip of the fang to the apex of the first tooth being usually ij times, or less, or at any rate not greater than if times the distance between the apices of the two large teeth : mesial surface of both upper and lower jaw usually with fairly strong development of feather bristles. In the male, the tibia of the palp is scopulate inferiorly over some portion of its length: upper margin of basal enlargement of flagellum blunt or sharp, but not greatly compressed into a thin high keel (except perhaps in darlingi) : no keel on the mesial surface of the fang just in front of the anterior bend of the flagellum. Posterior median angles of the two halves of the first abdominal sternite in the adult female either rounded or rectangular, never produced into narrow lobes. Posterior legs usually with sparsely disposed long hairs (occasionally as in sericea with nume- rous very long ones), never forming a distinct mane in either sex. (Species mostly nocturnal in habit and plainly coloured, being uniformly yellow or brown: only occasionally striped with black, or with strongly con- trasting colours on the abdomen.) A. Flagellum slightly produced anteriorly, the anterior bend situated fairly far forwards, over the first or second tooth: shaft long and subcylindrical over the greater portion of its length and with a projecting tooth near to the apex, or distinctly bifurcated distally. 1. Flagellum very long, reaching backwards as far as the ocular tubercle- or a little further, strongly sinuate not far from the apex and just proximal to its lateral tooth which is short and sharp : the shaft of the flagellum is low lying, being only just above the basal enlargement anteriorly. 5. lethalis typicus Koch. 2. Similar thereto, but shaft of flagellum straight, not sinuate, the shorter apical branch ending bluntly and its surface minutely serrulated. 5. lethalis rectus var. nov. 3. Flagellum reaching backwards beyond the middle of the mandible but terminating some distance in- front of the eye tubercle, only lightly sinuate between the apex and the short sharp lateral tooth; the apex laterally com- pressed and dilated above into a knife-like edge: anteriorly, the recurrent portion almost touches the basal enlargement. 5. venosa Purcell. 4. Flagellum reaching backwards beyond the middle of the mandible but not reaching the head-plate, terminating in a short sharp subulate apex at the base of which a fine straight pallid bristle-like process springs from the upper edge and extends backwards beyond the apex: proximally the shaft lies close to the basal enlargement, which is well elevated: anterior bend of flagellum situated behind the level of the first tooth and nearly over the second. Lower jaw with two or three small intermediate teeth. Dorsal surfaces of body, head-plate, and appendages strongly infuscated throughout in both sexes. 5. fusca C. L. Koch. 5. Similar to fusca, but the spine-like process near the apex of the flagellum much shorter, not extending as far as the apex: anterior bend of flagellum situated almost immediately over the first tooth. Lower jaw with only one intermediate tooth. Dorsal surfaces somewhat infuscated. 5. toppini Hirst. Annals of the Transvaal Museum 45 6. Flagellum not quite reaching the head-plate: towards the apex it presents a distinct bend and then bifurcates into two sickle-shaped processes, each finely pointed at the tip, the mesial and more slender process carrying microscopic setose teeth on its surface. 5. furcifera Kraep. B. Recurrent portion of flagellum long, usually reaching as far as the ocular tubercle or beyond, but not toothed nor bifurcate at the apex, nor serrated along its upper edge (or only faintly serrated immediately anterior to the distal sinus in Venator). The shaft is broadest at the anterior bend but is not strongly flattened there. 7. Flagellum strongly sinuate near to the apex, proximally tying close above the basal enlargement which is comparatively short and its upper edge arched : the anterior bend lies over the apex of the first tooth or a little anterior thereto. Only one small tooth between the second and third large ones of the upper jaw, or if two are present one of them is quite minute. S. venatov Poc. 8. Flagellum reaching back well beyond the ocular tubercle, straight throughout, or slightly sinuate towards the apex, the proximal portion situated well above the basal enlargement which is very long and low, the upper edge thereof being straight for quite a long distance. Usually two rather small teeth between the second and third large ones of the upper jaw, but the first of these may be quite minute or practically absent. A very strong develop- ment of long and straight stout spines on the outer and upper surfaces of the upper jaw. - 5. monteiroi Poc. C. Procurrent portion of flagellum short: shaft long and more or less cylindrical, at any rate over the first half of its length, some- times a little flattened, finely serrated for some distance along the upper or lower edge distally or along a curved or spiral ridge (in spiralicornis the serrated edge may be very short or perhaps obsolete). 9. Terminal fang of upper jaw very short, with fairly strong blunt internal tooth dorsally against which the lower jaw closes : shaft of flagellum somewhat flattened, more especially at the anterior bend which is rather broad, reaching back a little beyond the middle of the mandible, lightly arched in the distal third and gradually tapering to a fine point. Anterior bend of flagellum over the first tooth, or the interval between the second and first teeth. A fairly strong development of long pointed spines on upper surface of chelicera. 5. schlechteri Purcell. Terminal fang of upper jaw quite long, with a minute internal tooth. 10. Flagellum reaching to a point between the middle of the mandible and the ocular tubercle, bearing a more or less distinct denticulate crest in the posterior third or fourth of its length: there are usually two light curves in its course, a long one proximally and a short one distally (but these are sometimes absolete), and between them is a shallow dorsal sinus situated in the distal third. Viewed from above, the distal portion of the flagellum beyond the sinus is practically in a line with the main axis, and the shaft is not abruptly narrowed at the sinus, but tapers gradually from this region to the apex. Anterior bend of flagellum situated over the second tooth. Basal 46 Annals of the Transvaal Museum enlargement high, not much elongated, the anterior upper margin being rather lightly curved, but not forming an angle with the posterior ridge which is strongly curved. 5. serraticornis Purcell. 11. Similar thereto, but flagellum reaching as far as the ocular tubercle. A dense group of long strong spines on the upper and outer surfaces of the chelicera. S. serraticornis var. umtalica Hewitt. 12. Flagellum reaching just beyond the middle of the mandible, spirally twisted, and carrying a spiral serrated crest over the distal two-thirds of its length except near to the tip. Basal enlargement short, with curved upper margin. On the upper surface of the chelicerae there are a few long slender spines and stiff bristles but practical^ no stout spines. 5. strepsiceros Kraepelin. 13. Somewhat like serraticornis, but the denticulate crest on the flagellum feebly developed, being restricted to the inferior edge at the distal sinus. The flagellum is twisted as well as curved at the well-marked distal sinus : viewed from above, the portion beyond the sinus is quite straight, or nearly so, and parallel with the main portion of the shaft but distinctly external thereto: at the distal end of the sinus, the shaft becomes rather suddenly reduced in thickness, and thence to the apex is gradually drawn out to a fine point. Basal enlargement somewhat elongated, the upper margin being straight or nearly so for a considerable distance: shaft rather broad at the anterior bend, but otherwise not flattened. On the lateral and upper parts of the chelicerae there is a very strong development of straight stout spines. 5. spiralicornis Purcell. D. Recurrent portion of flagellum very short, not or scarcely extending back beyond the basal enlargement, and not strongly flattened over the greater portion of its length : procurrent portion very short, the anterior bend situated over the second tooth, or the interval between the first and second: on the inner dorsal edge of the upper jaw, just in front of the anterior bend of the flagellum, there is a small or minute tooth. 14. Flagellum an upstanding process, shaped somewhat like the horn of a rhinoceros, broad and strongly flattened from front to back at the anterior bend, tapering gradually towards the apex which is truncate, the central core projecting out therefrom a little: tip of flagellum situated just above the middle point of the basal enlargement. 5. schonlandi Pocock. 15. Flagellum somewhat similar, but at the apex is an enlarged globose swelling, the cuticle of which is minutely pubescent, thin, and pale, thus markedly contrasting with the shaft : the shaft is continued on the inner side of the swelling into a sharp-pointed flanking spine : the shaft is short and stout, being very broad at the anterior bend where it is flattened from front to back. Long spines and stout bristles are fairly well developed on the upper surface of chelicerae. 5. globicornis Kraep. E. Flagellum not, or only slightly, produced anteriorly, the anterior bend overlying the second tooth ; recurrent portion short, strongly flattened over the greater portion of its length: upper jaw on its inner dorsal edge, near to the anterior bend of the flagellum, furnished with a fairly conspicuous tooth, or two small teeth, against which the lower fang closes. Annals of the Transvaal Museum 47 1 6. Shaft of flagellum extending backwards about as far as the basal enlargement: at the anterior bend it presents the appearance o.f a broad stiff membrane: this quickly becomes much broader, but at the same time is strongly folded longitudinally along the middle, so that a transverse section of the flagellum at about the middle point of its length would be V-shaped, the angle being directed upwards and forming the upper margin of the flagellum in side view. In the apical third, the membrane tapers gradually, the apex being not very sharp, and at a short distance therefrom on the inner side is a strong accessory tooth. The single dorsal tooth on the inner side of the fang projects forwards and is long and strong. 5. ferox Pocock. 17. Shaft of flagellum reaching back to a point which is a little further from the hind margin of the basal enlargement than this from the anterior bend. Tip of flagellum pointed and barbed somewhat like an arrowhead: the terminal portion is also armed at the base with two strong serrate teeth, one external and the other inferior. 5. sagittavia Pocock. 18. Tip of flagellum nearer to the posterior margin of the basal enlarge- ment than is that margin to the anterior bend. Apex blunt. At the anterior bend, the shaft is very broad, becoming strongly twisted as it passes backwards : not far from the apex, its upper edge presents a small sharp tooth. On the inner dorsal edge of the fang of the upper jaw, there are two small tubercles near to the anterior bend of the flagellum. 5. suffusca Hewitt. F. Like D, but flagellum a little longer and the anterior bend situated further back over the intermediate small teeth. 19. Recurrent portion of flagellum reaching back a little beyond the posterior border of the basal enlargement, and becoming gradually attenuated at the tip : basal enlargement with a high crest and upright posterior border. 5. darlingi Poc. 20. Recurrent portion a little longer than in darlingi, running backwards and upwards at an angle of 450, straight throughout the greater portion of its length but distinctly sinuate in the distal fourth, the apex being directed upwards : basal lamina not so elevated, its outline about semicircular. Head- plate, mandibles and abdomen with black stripes. S. sericea Poc. 21. Flagellum still longer, gradually tapering to a slender apex, which is situated about midway between the anterior bend and the ocular tubercle; shaft doubly sinuate from side to side and strongly curved in a vertical plane in its distal half, the apex being directed downwards. Basal enlargement high and almost circular in outline. 5. alstoni Purcell. 22. Shaft of flagellum strongly flattened over the first portion of its length, then becoming abruptly narrowed distally so that the apical third or fourth is setiform : total length of recurrent portion equal to about twice the distance between the anterior bend and the hind margin of the basal enlargement, which is rounded in outline. First tooth of lower jaw situated midway between the second tooth and the tip of the fang: lower jaw with feathered bristles on its inner surface. Upper jaw with no distinct tooth on the superior inner margin near the base of the fang. Dorsal surfaces with blackish stripes. 5. celeripes Hirst. G. Shaft of flagellum short, directed forwards and extending about as far as the tip of the fang or even a little anterior thereto, 48 Annals of the Transvaal Museum then curving upwards a little but not or only slightly recurved, broad and strongly flattened throughout, distally with several short lobes or processes. In the upper jaw, the single series includes two anterior strong teeth, followed by a short interval, and then two or three teeth. 23. Flagellum terminating in a short sharp strongly curved spur, directed outwards and upwards, and composed of two closely appressed pieces : on the outer edge of the shaft, near to the apex, is an obtuse lobe- like process, and on the inner edge, still nearer to the apex, is a curved filiform process minutely serrulated along its outer edge. 5. cervina Purcell. 24. Similar thereto, but flagellum terminating in a broadly ovate lobe, with rounded apex, and provided with a straight acute process extending outwards from the posterior upper part: the upper edges of the lobe and its process are continuouslv serrated. 5. collinita Purcell. 23. Shaft of flagellum broadening out distally and dividing into three divergent backwardly directed slender processes, the middle one of which is shortest and stoutest, being straight, indurated and sharply pointed at the tip, the other two being curved and comparatively weak, the shorter mesial one being finely serrulate above near its apex. 5. alcicornis Kraepelin. Group II. Dentition of the upper jaw very dissimilar in the two sexes; in the female continuous, and closely resembling that of Group I; but in the adult male with a long toothless space in the middle of the single series (except in Junodi), thus separating the teeth into two groups of two teeth each, the teeth of the distal group varying much in size and occasionally with an additional small tooth anteriorly. Adult female with the first abdominal sternite more or less produced at the hind angles mesially into a pair of narrowed lobes (cp. PI. Ill, fig. 11). The species are all diurnal, and at once distinguished from the nocturnal species of group I by the strongly contrasting colours of the abdomen, the sides of which are white or yellow, whilst the posterior tergites are black or strongly infuscated, the anterior tergites being paler mesially but usually becoming infuscated laterally, thus presenting a dark lateral band on each side which gradually merges with the blackened area of the posterior tergites. The upper margin of the basal enlargement of the flagellum of the male is elevated into a sharp keel: the lower jaw usually has a very long fang, and on the outer side distally is a sharp keel, which is sometimes continued as a prominent granular crest towards the base of the jaw: feather bristles are poorly developed on the inner surface of the lower jaw, and are not very well developed on the upper jaw. Posterior legs usually with numerous long hairs which are some- times arranged in a distinct mane in adult males. A. Terminal fang of upper jaw without a long keel on its mesial side, but provided with a dorsal forwardly projecting tooth just in front of the anterior bend of the flagellum, or with an abbreviated Annals of the Transvaal Museum 49 keel. Flagellum strongly flattened from front to back at the anterior bend, being as wide as the fang itself at this point, or nearly so. Shaft of flagellum filiform throughout. 26. Flagellum reaching back a little beyond the ocular tubercle and slightly blunted at the tip : the anterior bend situated far forwards, overlying the most distal tooth: basal enlargement very high, the whole upper margin elevated as a high laminar keel. On the inner side of the upper jaw, near to the anterior bend, is a distinct dorsal tooth pointing forwards. Two distal teeth of upper jaw moderately large, and in front of these is a more or less distinct though minute additional tooth. S. lateralis Koch. 27. Anterior bend of flagellum overlying the tooth preceding the long interval, the shaft terminating a little posterior to the hind margin of the basal enlargement. Three distal teeth in the upper jaw, the second and third being of moderate size. S. erythronotoides sp. nov. 28. Anterior bend more posteriorly situated, overlying the toothless interval: shaft reaching back about half way along the mandible. Basal enlargement high. Two distal teeth of upper jaw moderately well developed. S. erythronota Kraepelin. B. Shaft of flagellum not simply filiform, but more or less flattened and expanded in the distal portion, where the edges are frayed out into processes or deeply serrated. 29. Terminal fang of upper jaw with a dorsal tooth but no keel or only a very weak one on its mesial side. Flagellum broad at the anterior bend, which overlies the tooth preceding the long interval, reaching back about as far as the ocular tubercle, and filiform over the greater portion of its length, but in the terminal third it expands slightly into a lamina and divides into two main portions, the longer one being drawn out to a filament with serrated margins, the shorter one ending abruptly but bearing a short and slender serrated extension running parallel to the longer process just mentioned. Two distal teeth of upper jaw moderately large. S. maraisi Hewitt. 30. Anterior bend of flagellum overlying the toothless interval, the shaft reaching back beyond the middle of the mandible, being deeply and finely serrated along its upper edge in the distal half which is much flattened from the sides. Fang of upper jaw with an exceptionally high keel arising from the upper surface along its inner edge, just above the distal group of teeth and in front of the flagellum. S. spectralis Purcell. 31. Anterior bend situated as in spectralis , the shaft short, about as long as the terminal fang of the lower jaw, and serrated along three edges in its distal half which is expanded except towards the apex. A thin upright blade- like keel, serrated anteriorly, and provided with a sharp forwardly directed tooth about the midde of its length, occurs on the inner side of the fang of the upper jaw. 5. vincta Koch. C. Terminal fang of upper jaw grooved on its mesial side superiorly for the reception of the procurrent portion of the shaft of the flagellum, the groove continued in front of the anterior bend, being bounded mesially by a long keel which extends from the anterior bend towards the tip of the fang. Flagellum not broad at the anterior bend, the shaft filiform, not expanded. a. Anterior bend of flagellum posteriorly situated, over- lying some portion of the toothless interval. 4 50 Annals of the Transvaal Museum 32. The distance from the tip of the flagellum to the hind margin of the basal enlargement is less than the length of the basal enlargement, which is much longer than high. Two distal teeth of upper jaw weak. In the lower jaw, the distance from the tip of the fang to the apex of the first tooth is about equal to ij-if times the distance between the apices of the two large teeth. 5. coquinae typicus Hewitt. 33. Similar thereto, but fang of lower jaw longer, the distance from the tip of the fang to the apex of the first tooth being equal to about 2-2 \ times the distance between the apices of the two large teeth. 5. coquinae orangicus var. nov. 34. Flagellum short, reaching back only a little beyond the hind border of the basal enlargement, and suddenly expanded at its apex into a slender funnel with obliquely truncated margins. Two distal teeth of upper jaw of moderate size, the keel on the inner side of the fang strong and outstanding. 5. tubicen Krapp. b. Anterior bend of flagellum more anteriorly situated, overlying the tooth preceding the long interval, or a trifle anterior thereto. 35. Flagellum reaching back nearly to the end of the mandible: basal enlargement much longer than high: two distal teeth of upper jaw weak. 5. derbiana Pocock. 36. Recurrent portion of flagellum short, ending in a fine point just posterior to the basal enlargement which has a high dorsal keel. The toothless interval of the upper jaw is practically obsolete as such, the two distal teeth being large, strongly compressed and sublaminate. The teeth of the lower jaw are close together near the base, the fang being very long. 5. junodi Purcell. c. Anterior bend of flagellum still more anteriorly situated, overlying either the most distal tooth or even in front thereof. 37. Flagellum slender, reaching to the ocular tubercle, sharp-edged near the apex. Two distal teeth of upper jaw usually weak, but the second may be of moderate size. Viewed from the side, the terminal fang of the upper jaw is directed obliquely upwards to a slight extent, being at an obtuse angle with the main axis of the jaw. 5. hostilis White. 38. Flagellum terminating in a slightly upcurled point above the ocular tubercle, the apex being very finely pointed but not cultrate. The shaft is distinctly flattened over a considerable portion of its length. Viewed from the side, the short terminal fang of the upper jaw is directed horizontally forwards in a line with the rest of the jaw, which is itself greatly produced in the region of the extended toothless interval. Two distal teeth of upper jaw of moderate size. 5. mavshalli Pocock. 39. Flagellum not reaching so far as the hind margin of the chelicera, its apex presenting a sharp cutting edge for a short distance dorsally. Toothless interval comparatively short, the distance from the apex of the second tooth to the tip of the fang, being subequal to the distance between the apices of the second and fourth teeth. 5. tookei sp. nov. 40. Flagellum reaching to a point slightly posterior to the ocular tubercle, the distal fifth being bent strongly downwards, the apex minutely forked, and the indurated outer upper edge of the bent portion is notched at a short Annals of the Transvaal Museum 5i distance from the apex. Two distal teeth of upper jaw large, the apex of the second tooth being just a trifle nearer to the apex of the fourth than to the tip of the fang, or midway between the two. 5. bechuanica Hewitt. 41. Flagellum reaching to the ocular tubercle, lightly sinuate in its distal fifth, flattened from above near the apex and notched on the inner edge. First tooth of upper jaw rudimentary, separated by an interval from the second, which is large and posteriorly situated, so much so that the distance from the apex of this tooth to the tip of the fang is much greater than the distance between the apices of the second and fourth teeth. 5. hamata Hewitt. Group III. Related to II C, but upper jaw with two rather long toothless spaces in the single series of teeth. A very well developed mane present on the hind legs. Tarsus II with a row of four distinct spines on the outer side superiorly. 42. Flagellum with the anterior bend about in a line with the mid-point of the first toothless space, the recurrent portion more or less straight throughout, about reaching to the hind edge of the chelicera, the posterior margin of the basal enlargement being decidedly anterior to the mid-point of the length of the flagellum. S. chelicornis Licht. a. A dense pile of short pale hairs on the upper surfaces of the patella and tibia of the palp in the male. In the upper jaw, the distance between the first and second teeth is decidedly greater than the distance between the second and fourth. S. chelicornis pubescens var. nov. b. Similar thereto, but without a pile of short pale hairs on the upper surfaces of the palp. 5. chelicornis rufescens var. nov. c. In the upper jaw, the distance between the first and second teeth is less than the distance between the second and fourth. 5. chelicornis macrognathus var. nov. 43. Flagellum reaching back to a point about midway between the ocular tubercle and the hind margin of the basal enlargement, which is about in a vertical line with the mid-point of the flagellum; the anterior bend is rather more posteriorly situated, and above it the shaft makes a bold high curve (not depressed as in chelicornis ). 5. villosa Purcell. Group IV. The single series of teeth in the upper jaw presents no long toothless intervals, and comprises six teeth (or only five when the normally minute fourth tooth is obsolete or lacking), of which only the one basally situated is of large size, the first tooth being small, the third also small, and the second still smaller. In the lower jaw, the apex of the first tooth is nearer to the tip of the fang than to the apex of the basal large tooth. Anterior bend of flagellum far forwards, in front of the first tooth, the procurrent portion of the shaft not sunk into a groove of the upper jaw, so that there is no keel nor tooth on the mesial surface of the terminal fang. Tibia of 4—2 52 Annals of the Transvaal Museum palp well scopulate inferiorly, the whole palp being subequal to or only very slightly shorter than leg III. Females also have six teeth in the single series of the upper jaw, the first, third and sixth being of large size. 44. Flagellum reaching back almost to the ocular tubercle, in its distal fourth flattened dorso-ventrally and ending in a flat lance-like apex. S. hastata Kraepelin. 45. Similar thereto, but the terminal fourth of the flagellum is deeply grooved above and for a short distance keeled below. 5. ornithorhyncha Hewitt. 46. Flagellum much shorter, presenting a large bold curve at the anterior bend, and directed obliquely upwards in a sinuous line with two curves in its course, a short deeper one near the apex, and a longer shallower one proximally : it reaches backwards only a little beyond the hind margin of the basal lamina, which is remarkable in being produced upwards as a short outstanding process. 5. methueni Hewitt. Group V. Dentition of upper jaw similar in the two sexes, the single series comprising a distal tooth at some distance from the tip of the fang, followed by 1-3 small teeth and one large one, then 3-5 small teeth, and finally another large one. Terminal fang of upper jaw with a fairly long high ridge on its mesial side, starting near the anterior bend of the flagellum and proceeding towards the tip of the fang. Terminal fang of lower jaw short, the distance from the tip thereof to the apex of the first tooth being only about 1 J times as long as the distance between the apices of the two teeth. Tibia of palp not scopulate below, the whole palp being decidedly shorter than leg III. 47. Flagellum reaching back a little behind the middle of the mandible, flattened out in its distal half which is slightly twisted and fringed along the edges: anterior bend about on a level with the second tooth. Upper surfaces with black stripes. S. line ala Koch. Genus Zeriassa Pocock. Zeriassa cuneicornis Purcell, 1899. Annals S. Af. Mus. 1. p. 413, figs. 20 and 20 a. The type, described under the name of Solpuga cuneicornis , came from Southern Rhodesia. Zeriassa purcelli Hewitt [PI. VII, fig. 36], 1914. Annals Transvaal Mus. iv. p. 163, fig. 25. This species is only known from Newington, N.E. Transvaal. The female referred to in the original description does not belong to the species. A small exampLe from Newington, with the two long preocular spines characteristic of the genus, is no doubt the true female of this species. In this specimen, the terminal fang of the upper jaw is comparatively long, the first tooth is small, the second tooth is very long and large, being the longest in the whole dental series, the third is small, fourth of moderate size, and the fifth large: this may be regarded as completing the single series, although the outer Annals of the Transvaal Museum 53 row of the double series is quite continuous with the single series. The outer row includes four teeth of which the distal one is the largest, but is not so large as the adjoining fifth tooth of the single series. The inner row includes three teeth, of which the first is largest and is widely separated from the second as well as from the single series: the basal tooth is quite small. In the lower jaw there is only one intermediate tooth. Both upper and lower jaw are strongly compressed from side to side : the lateral distal keel on the outer side of the lower jaw thus becomes carried ventralwards, and is not continued into the granular crest which runs along the middle of the basal portion of the jaw. The lower jaw has great depth, and thus in side view has a massive appearance. Stridulatory ridges are well developed on the chelicerae, number- ing about 12 in the male, and 9-10 in the female. Feather bristles are not well developed on the lower jaw of the male: it resembles that of the female but is not so strong. The flagellum of the male has various points in common with that of cuneicovnis , and it may be noted that the upper margin of the basal enlargement is elongated in both, and that the upper and posterior margins are acutely inclined to each other, with a large high keel at the angle. The procurrent portion of the shaft lies on the upper surface of the jaw, and is not sunk into a groove: on the mesial side of the fang superiorly, just in front of the anterior bend, an ill defined ridge occurs. Key to the S. African species of Zeriassa. 1. Shaft of flagellum short, very broad at the anterior bend, being there almost twice the width of the fang below it, thence gradually narrowing towards the middle where a slight twist occurs and the shaft bends outwards slightly, ending in a point (apparently) a short distance beyond the basal enlargement. Basal enlargement with triangular out- line when viewed from the side. Dentition of single series of upper jaw continuous, including five teeth: lower jaw with one intermediate tooth. Z. cuneicovnis Purcell. 2. Shaft of flagellum very broad throughout its length, tapering somewhat up to the distal twist but expanded a little near the apex, which carries a short slender hook-like process: it extends only a trifle beyond the basal enlargement. Upper jaw with six teeth in the single series: lower jaw with two intermediate teeth, and the distance from the tip of the fang to the apex of the first tooth is subequal to the distance between the apices of the two large teeth. Z. purcelli Hewitt. Genus Daesia Karsch . Daesia suhulata Purcell, 1899. Annals S. Af. Mus. 1. p. 393, fig. 12. The type male came from Van Wijk’s Vlei, Carnarvon. Daesia hernhardi Pocock, 1900. Annals Mag. Nat. Hist. 7, vi. p. 317. The type is a female from Hex River Valley. The character of spines and ridges on the sides of the abdomen interiorly, considered by Pocock to be distinctive of this species, occurs also in leipoldti, lineata, and probably also in other species. I may remark that these structures are easily overlooked, for the sides of the abdomen are densely hairy and the spines and ridges quite small. To see them it is necessary to stretch out the integument and examine 54 Annals of the Transvaal Museum under the low power of a compound microscope. The ridges mentioned by Pocock (7) occur in the soft skin of the sides inferiorly, immediately lateral to the interval between sternites III and IV. The spines on the soft skin of various segments, as specified by Purcell (12) who described them as “ claw- like,” are all quite slender. Daesia namaqua Kraepelin, 1899. Das Tierreich, p. 94, fig. 63. The type came from Gt Namaqualand, and Kraepelin records it also from Steinkopf in Little Namaqualand. Daesia betschuanica Kraepelin, 1908. Denk. med. nat. Gesell. Jena, xm. p. 273. The type male came from Lobatsi. Daesia kolbei Purcell, 1899. Annals S. Af. Mus. 1. p. 391, fig. 10. This is founded on a female specimen from Bulawayo. Daesia pallida Purcell, 1899. Annals S. Af. Mus. 1. p. 382, fig. n. The type is a female specimen from the Kenhart div. C.P. Daesia leipoldti Purcell, 1899. Annals S. Af. Mus. 1. p. 388, fig. 9. This species is based on female specimens from the Clan william div. C.P. Daesia hottentotta Kraepelin, 1899. Das Tierreich, p. 95, fig. 64. According to Kraepelin, this is the commonest species of South West Africa, and he records it from various localities in Damaraland and Gt Nama- qualand. An example from Mt Temple, Bechuan aland (T. C. Lanham) is perhaps referable to this species. It has a flagellum like that of lineata, and along the middle is a dark midrib. The dentition of the upper jaw is not very different from that of lineata : the single series includes four teeth, of which the first and third are of moderate size, being not much smaller than the second and fourth, and the inferior outline of the fang is curved, not straight. The same form occurs at TJpington (Miss Lennox). Daesia lineata Pocock [Pis. VI, figs. 27, 28 and 31, and VIII, fig. 43], 1902. Ann. Mag. Nat. Hist. 7, x. p. 7, PI. II, figs. 3, 3 a and 3 b. = D. schreineri Purcell, 1903, Annals S. Af. Mus. 111. p. 14, PI. 1, fig. 11. The type of this species was regarded as representing a distinct genus, and was described by Pocock under the name of Broomiella lineata. I am satisfied that Pocock’s specimen was an aberrant example of the same species as was described subsequently by Purcell as D. schreineri. The most noteworthy feature of Pocock’s specimen, according to the figure, is the small size of the fourth tarsus, which apparently is shorter than the third, and even the claw of the fourth is represented as shorter than that of the third tarsus. Now, in the genus Daesia, the claw of the fourth leg is normally much longer than that of the third, and so also is the tarsus as a whole, but occasionally excep- tions are found. The exceptions observed by me are without doubt referable to the specifes D. schreineri, agreeing with normal examples of that species in the dentition and in the flagellum, and indeed are only remarkable in the characters of the fourth leg. In an adult male from Hopetown tarsus IV of the right side is normal, but the left tarsus IV is two-jointed, the whole leg being shorter than that of the right side, although its basal segments bear five malleoli : the tarsal claws of the shorter leg are noticeably short, being shorter than those of tarsus III. and the tarsus as a whole is only a little longer than Annals of the Transvaal Museum 55 tarsus III. Again, a small female example from De Aar, where typical examples of schreineri occur, has the left fourth leg normal throughout, but the right leg is short and its tarsus two-jointed with quite short claws: there are five malleoli as usual. The peculiarities of these aberrant specimens are perhaps the result of injury during earlier stages in the life of the individual. It should be noted that Pocock’s figure of the flagellum does not represent that organ so correctly as the figure accompanying the description of schreineri, assuming the identity of the two species. The type of lineata came from Pearston : that of schreineri from Hanover. We have the species from De Aar (S. C. Cronwright-Schreiner and Miss E. Friedlander), Tafelberg (Miss A. Gadd), Longhope (Miss E. Abrahamson), Zandkraal near Steytlerville (Miss A. Geard), Hopetown (Miss Stegmann), Smithfield (Dr R. Broom), and female specimens apparently referable thereto from Klerksdale near Middelburg (B. Marais). The stridulatory ridges of the chelicerae are decidedly longer than in Solpuga, but are only 4-6 in number: the bristles on the mesial surface are much as in Solpuga and the feather bristles towards the distal end of the series are not enlarged. The lower jaw has feathered bristles on its mesial surface: the distance from the tip of the fang to the apex of the first tooth is about equal to i| times the distance between the apices of the two large teeth: the intermediate tooth may be present or absent in male specimens from the same locality. In side view, the lower margin of the terminal fang of the upper jaw is quite straight: this seems to be a very constant character. Rotated forwards, the flagellum either just reaches the tip of the fang, or slightly surpasses, or scarcely reaches the tip: during life this probably can be rotated through an angle of 180°, or slightly more, but not very much more. The tarsus of the palp is slightly movable in both sexes: this is not the case in Blossia. Daesia rhodesiana Hewitt, 1913. Annals Transvaal Mus. iv. p. 165, fig. 26. The type male came from Lundi River, S.E. Rhodesia. Daesia schultzei Kraepelin, 1908. Denk. med. nat. Gesell. Jena, xiii. p. 274, figs. 5 and 6. The type is an adult male from the Kalahari. Key to the S. African species of Daesia Karsch. 1. Upper jaw with a very long slender toothless terminal fang. Teeth of single series absent. Flagellum drawn out behind into a straight subulate shaft which exceeds the basal disc in length. D. subulata Purcell. 2. Terminal fang of upper jaw rather long, followed by the single series of four teeth — two fairly large ones in front, then a small one, and then a large tooth. Flagellum drawn out into a shaft which is longer than the disc. D. namaqua Kraepelin. 3. Terminal fang of upper jaw of moderate length or short, the single series of teeth continuous ; the flagellum more or less elongated and pear- shaped, without a long shaft, though drawn out a little distally. 56 Annals of the Transvaal Museum a. The single series of teeth in the upper jaw comprises two large ones in front, one small intermediate tooth, followed by a large tooth. D. hottentotta Kraepelin. b. Dentition similar to that of hottentotta but two small intermediate teeth in the single series. D. pearsoni Hewitt. c. The single series of teeth in the upper jaw comprising one small one in front, then a large one, then a small intermediate tooth, and finally a large one. D. lineata Pocock. 4. Terminal fang of moderate length. Flagellum attenuated distally but not produced into a long shaft. Basal tooth of lower jaw with a broad more or less truncated apex. In the upper jaw, the longest and largest tooth is separated from the distal tooth by a deep bay : a. This bay including one small tooth. D. betschuanica Kraepelin. b. The bay being quite toothless. D. rhodesiana Hewitt. 5. Terminal fang of upper jaw specially long, followed by a single series comprising only two teeth, both of which however are large. Flagellum continued beyond the capsule into a short dorsal process and a longer ventral one, connected together by a fold of membrane, the latter process being less than half the length of the capsule. D. schultzei Kraepelin. Genus Blossia Simon. Blossia setifera Pocock, 1900. Ann. Mag. Nat. Hist. 7, v. p. 301, fig. 6. The type male came from Salisbury, Mashonaland. Blossia namaquensis Purcell, 1901. Annals S. Af. Mus. n. p. 211, fig. 2. The type male came from Steinkopf, Namaqualand, C.P. Blossia unguicornis Purcell [Pis. VII, figs. 39 and 40, and VIII, figs. 44 and 46], 1901. Annals S. Af. Mus. n. p. 213, fig. 3. Types from Dunbrody, Uitenhage div., and the species is also known to me from Alicedale (F. Cruden), Somerset East (E. Driver), and Linedrift, Peddie (B. Marais). In the male the head-plate is beset with short spinules, but each eye is protected on its mesial side by a curved row of upstanding spines: the chelicerae have numerous, rather short, sharp pointed, stout spines and a few long ones superiorly, and the thoracic tergites are also fringed with pointed spines: abdominal tergites with very short spines and weak spinules, the posterior one or two tergites with slender setae which are slightly notched at the tips: a few spines occur on the upper surfaces of the trochanters and femora of legs II-IV. In the females of unguicornis , and probably of other species also, the head- plate is armed with short spinules or prickles, and the tergites and chelicerae with spines : these spines are not so strong nor so numerous as those of the male, the first tergite for example having a single row of long slender spines in the female, but far more numerous, shorter and much stouter spines, constituting more than one row, in the adult male. Annals of the Transvaal Museum 57 Female characters are by no means constant throughout the genus Blossia. In a female from N.W. Gordonia, the tergites and chelicerae are armed with long and rather stiff setae, but not with spines, and the head-plate is devoid of spinules: moreover, the jaws are more slender and longer than in females of unguicornis. In this genus, modified fleshy hairs on the sternites are not confined to the adult males, nor to the second sternite, although most strongly developed on that sternite in males. In adult females of B. unguicornis there occurs a mesial papilla on both fourth and fifth sternites : this papilla is split down the middle, and carries on each side a long pinkish thickened hair like those on the second sternite of males, but not nearly so stout. They also occur on the same sternites in the adult male. Similar modified hairs occur in a female Blossia (cp. setifera Poc.) from Salisbury, being found on all three sternites II, IV and V: on II there are three or four modified hairs on each side, but in IV and V only two hairs occur on each side of the midline: they are situated rather further apart than in unguicornis. These modified hairs are probably homologous with the stigmatic combs of the Galeodidae: the combs of Galeodes occur on the second and third sternites, and modified setae in a corresponding position are found also on the fifth sternite. Blossia crepidulifera Purcell, 1901. Annals S. Af. Mus. n. p. 215, fig. 4. The type came from Robertson C.P., and Dr Purcell cited also several records in the Worcester div. Blossia karrooica Purcell, 1901. Annals S. Af. Mus. 11. p. 217, fig. 5. The types were taken at Prince Albert C.P. Blossia litoralis Purcell, 1903. Annals S. Af. Mus. in. p. 4, fig. 3. Types from the western shore of the Cape Peninsula. Blossia laminicomis Hewitt, 1919. Annals Transvaal Mus. vi. p. 65. The type was taken at De Aar, C.P. -» The point of origin of the flagellum is a little posterior and dorsal to the distal end of the row of feather bristles. Blossia falcif era Kraepelin [Text fig. 10 a\, 1908. Jena. Denkschr. xm. p. 277, fig- 8. This species is known from the neighbourhood of Windhuk : other localities cited are Neudamm and Tsumab. I have previously recorded the species from Quibis in Gt Namaland, and it is known to me also from Douglas C.P. (Dr R. Broom). These southern specimens do not agree absolutely with Kraepelin’s figure of the type, but the flagellum is very similar throughout, and I think it advisable to regard such differences as occur in the dentition as of varietal value only. The lower jaw of the type, which is figured with five teeth, instead of four as occur in the specimens examined by me, is perhaps abnormal, for, according to the description, the type specimen presents con- siderable differences in this respect in the jaws of the two sides. In the upper jaw of the type, the terminal fang is fairly large- the distance from the tip to that of the first tooth being however distinctly less than the distance between the apices of the fifst and fourth teeth. The example from Quibis 58 Annals of the Transvaal Museum is not very different, but the terminal fang is a trifle shorter, the distance from its tip to that of the first tooth being considerably less than the distance between the first and fourth teeth. The male from Douglas is markedly different from the above, the terminal fang of the upper jaw being longer and more slender: the distance between the tip of the fang and the apex of the first tooth is equal to, or very slightly greater than, the distance between the apices of the first and fourth teeth. This variety may be termed var. dolichognathus (Text fig. io b). The distal dorsal bristle in all these specimens reaches to a point about midway between the first tooth and the tip of the fang, or a trifle nearer to the first tooth: its surface is minutely prickled almost up to the pointed apex, which tapers gradually. On the outer side of the upper jaw there is an oblique row of simple bristles, the bases of which are in a line more or less parallel with the dental series. These bristles vary much in length and size, some being quite short, others long : the most distal one, arising from a point opposite to the base of the lamina, is markedly stouter than the rest, and its surface is minutely prickled. In the allied species B. filicornis mihi, this row of bristles Text fig. io, a and b. Blossia falcifera Kraepelin. (a) Upper jaw of male from Quibis, viewed from outer side, (b) Ditto of var. dolichognathus from Douglas. is much more uniform, the distal one being scarcely stouter than the second, and the first three have prickled surfaces. In both species, the slender apex of the flagellum is upcurled. Blossia fimbriata Kraepelin, 1914. Beit. z. Kenntnis d. Land- it. Susswasser fauna Deutsch-Sudwestaf vikas, p. 128, fig. 3. The locality of the type is simply “ Deutsch-Siidwestafrika.” It seems to be very distinct from any other species known from S. Africa. Blossia echinata Purcell, 1903. Annals S. Af. Mus. hi. p. 16, PI. I, fig. 10. The types came from Hanover C.P. Blossia maraisi Hewitt, 1915. Records Albany Mus. in. p. 70, fig. 1. The type male came from Peddie. Blossia tricolor Hewitt, 1913. Annals Transvaal Mus. iv. p. 156, fig. 20. The types were collected at Quibis in Gt Nam aland. Annals of the Transvaal Museum 59 Blossia clunigera Kraepelin, 1908. Jena. Denkschv. xm. p. 275, fig. 7. The type was taken at Steinkopf in Little Namaqualand. Blossia obscura Kraepelin, 1908. Jena. Denkschv. xm. p. 278. This is based on a female from Khakea in the Kalahari. Blossia Jilicornis Hewitt, 1913. Annals Transvaal Mus. iv. p. 158, figs. 21 and 22. The types were taken at Alt Wasserfall in Gt Namaland. Blossia laticosta Hewitt, 1919. Annals Transvaal Mus. vi. p. 64. The type was taken at Blauwkop in the Zoutpansberg dist. Key to the S. African species of the genus Blossia Simon. A. Flagellum long and slender, widest near the base, tapering more or less gradually to the apex, and when directed forwards extending considerably beyond the tip of the upper jaw: basally, it is membranous with in-curved edges forming a more or less inflated capsule, with a greatly elongated aperture on the side adjacent to the jaw: distally, this gradually becomes a flattened rod. a. On the lower margin of the fang of the upper jaw, between the tip of the fang and the first large pointed tooth, there is a tooth- like projection formed by a laterally compressed obtuse lamina. 1. Distal dorsal bristle with proximal half stout and minutely granular (except at the base), the distal half more slender, subulate, and smooth. Flagellum suddenly acuminate at the apex when seen from the side. B. karrooica Purcell. 2. Distal dorsal bristle hardly reaching so far as the first pointed tooth, slender and quite devoid of granulation in any portion of its length, being precisely similar to the other bristles of the neighbourhood. Distal portion of the flagellum more slender than that of karrooica, and not so suddenly acuminate at the apex. B. maraisi Hewitt. b. No pronounced tooth-like lamina between the tip of the upper jaw and the first large tooth (it is represented by a minute tubercle or slight convex dilatation in echinata). 3. Distal dorsal bristle short1, hardly reaching so far as the first tooth, its surface minutely granular. Basal half of flagellum of more or less uniform width, thence becoming gradually attenuated towards the apex where it is slightly hooked. B. setifera Pocock. 4. Distal dorsal bristle almost reaching to the tip of the fang, its surface minutely roughened in the distal half. Flagellum with a broad basal half and a rod-like distal half fairly sharply differentiated. Dorsal edge of upper jaw furnished with a small sharp tooth which is almost midway between the base of the distal dorsal bristle and the tip of the fang. B. echinata Purcell. 5. Distal dorsal bristle strong and boldly curved, reaching about as far as the anterior border of the second tooth, tapering more or less uniformly throughout its length, and not very strongly denticulated on any part of its surface. Flagellum very much broader towards the base than in the distal 1 This character is inferred from the figure of the type. 6o Annals of the Transvaal Museum portion, the whole structure being fairly regularly attenuated from the widened basal portion up to the apex. The first tooth of the upper jaw is much nearer to the second than to the tip of the fang. B. falcifera Kraepelin. 6. Distal dorsal bristle very stout in its basal two-thirds, where the surface is rather strongly denticulated except near the base; then becoming attenuated rather suddenly, the apical third being slender, smooth and taper- ing to a fine point. Flagellum quite slender, being only slightly expanded at the base, and the distal half being practically uniform in breadth throughout. The first tooth of the upper jaw is as near to the second as to the tip of the fang. B. filicovnis Hewitt. B. Flagellum short, not or scarcely surpassing the end of the fang when directed forwards, the distal portion being expanded into a widely open capsule, more or less oval or pear-shaped in outline, and the basal portion forming a distinct narrow stalk of attachment. 7. Distal portion of upper jaw with a long and straight upper margin, the first three teeth subequal in size, the third being followed by a faiily long straight interval, the fourth tooth being small. Distal dorsal bristle long and straight, minutely echinated m its distal half, which is as stout as the basal half. B. cvepidulifeva Purcell. 8. Distal portion of upper jaw curved towards the apex; the single series is continuous and includes two strong pointed teeth distally, followed by two small teeth : between the more distal pointed tooth and the tip of the fang, is an inner blunt laminate tooth paired with an outer more conical tooth. Distal dorsal bristle undifferentiated. B. litoralis Purcell. 9. Similar to crepidulifera, but the distal portion of the upper jaw is curved : the single series comprises two large distal teeth, a small third tooth, and a large fourth in a continuous series. B. clunigeva Kraepelin. C. Flagellum a more or less pear-shaped capsule, with a wide elongated opening over the distal half or two-thirds of its length, the basal funnel-shaped part narrowed but not definitely petiolate. 10. Distal part of flagellum broad, densely covered with shaggy hairs. Outer wall of flagellum furnished with a strong well-marked yellow axis which terminates distally in a short sharp claw. Second abdominal segment inferiorly carrying two pairs of sickle-shaped fleshy hairs situated near the midline, e B. unguicovnis Purcell. 11. Second abdominal segment inferiorly with two groups of three fleshy hairs near the midline. Flagellum without prickles or setae on its surface, although the distal edges are slightly frayed: the yellow thickened rib is very broad, and the amount of free membrane dorsal to it is much less than in unguicovnis. B. laticosta Hewitt. 12. Flagellum without a longitudinal thickened rib. Between the first large tooth of the upper jaw and the tip of the fang, there is a small obtuse tooth (which is absent in unguicovnis ). B. tvicolov Hewitt. D. Flagellum like that of C, but widest about the middle of its length and gradually tapering distally to a slender apex. Annals of the Transvaal Museum 6r 13. Lateral wall of flagellum furnished along its middle with a thick pale yellow longitudinal rib, extending throughout the whole length: surface of flagellum thickly studded with prickles right up to the apex. Distal dorsal bristle becoming gradually thinner towards the apex, and covered with very minute spinules in its distal part. Third tooth of upper jaw rudimentary. B. namaquensis Purcell. 14. Third tooth of upper jaw only a little smaller than the first or the second. Distal dorsal bristle precisely similar to the other bristles which occur on the upper and outer surfaces of the chelicerae, and with only a trace of short fine setose prickles on its distal portion. Surface of flagellum closely studded with minute triangular denticles except towards the base, and on the narrowed distal portion. B. laminicornis Hewitt. E. Flagellum long, reaching beyond the tip of the fang when directed forwards; not very broad in any part, nor greatly attenuated either towards the apex or the base ; the cavity of the slightly expanded capsule, which occupies the distal portion of the basal half of the flagellum, has a short oval aperture: the lancet-shaped distal half of the flagellum is lightly curved, and the membrane bordering it along the upper or anterior margin is very finely serrated from the capsule almost up to the apex. B. fimbriata Kraepelin.^ Genus Gluviopsis Kraepelin. Gluviopsis australis Purcell, 1901. Annals S. Af. Mus. n. p. 219, fig. 8. The type came from Styrkraal near the Orange River in the Kenhardt div. This is the only species described from S. Africa, but Kraepelin gives some notes on a female example of the genus from Swakopmund. The flagellum of australis — which is only known to me through the descrip- tion and figure — is more or less oblong in shape, tapering towards the base, strongly compressed and bent outwards distally, the apex being very broadly subtruncated and lacerated. There are four large teeth in the single series of the upper jaw. Genus Hemiblossia Kraepelin. Hemiblossia bouvieri Kraepelin, 1899. Das Tierreich, p. 104, fig. 77. This is based on a single female specimen from the upper Zambesi. Hemiblossia O’neili Purcell [Pis. Ill, fig. 5 and VIII, fig. 42], 1901. Annals S. Af. Mus. 11. p. 217, figs. 6 and 7. This is based on several males and females from Dunbrody on the Sundays River. Also known to me from Alicedale (F. Cruden), De Aar (S. C. Cronwright- Schreiner) and Kimberley (Bro. J. H. Power). A figure of the flagellum of Alicedale specimens is given by me in Records Albany Mus. vol. 111. p. 12. The inner surface of the chelicera in both sexes has stridulatory ridges, simple bristles, and feathered bristles, as usual. 62 Annals of the Transvaal Museum Hemiblossia kalaharica Kraepelin, 1908. Denk. med. nat. Gesell. Jena, xm, p. 279. The type is a female from Khakhea in the Kalahari. Hemiblossia idioceras Hewitt, 1917. Ann. Natal Mus. 111. p. 687, fig. 1 a-c. This is based on an adult male from Kimberley. The species seems well separated from O’neili in the characters of the flagellum, and it is of interest to note that both forms have been taken at Kimberley. Mr Claude Fuller has taken this species at Pienaar’s Riv. Transvaal: it occurred in numbers within the nest of a Eutermes. Mr J. H. Power also found adult males in a termites’ nest at Kimberley, Nov. 1918. Key to the species of Hemiblossia Kraepelin. 1. Dorsal margin of flagellum rather strongly curved, though not quite semicircular, and at its distal end is an independent process which is slender, curved and hairy: the distal margin of the flagellum is some- what flattened out, but not into a distinct foliaceous appendage. H. O’neili Purcell. 2. Dorsal margin of flagellum lightly curved, distally with an acutely pointed termination : distal margin of flagellum flattened out on the side adjacent to the chelicera into a foliaceous extension the margins of which are fringed with hairs. H. idioceras Hewitt. Genus Melanoblossia Purcell. Melanoblossia braunsi Purcell, 1903. Annals S. Af. Mus. in. p. 6, figs. 4 and 5. The type came from Willowmore, and a juvenile specimen was taken at Hanover. Melanoblossia globiceps Purcell, 1903. Annals S. Af. Mus. in. p. 8, fig. 6. Type from O’okiep in Little Namaqualand. Key to the species of Melanoblossia Purcell. 1. “ Flagellum ” inconspicuous, being a straight hairy rod, like the setae immediately below it. Head-plate moderately convex, with notched setae of varying length, but not densely clothed with minute cylindrical bristles. Six small teeth in the single series of the upper jaw. Second abdominal sternite with two pairs of long narrow pointed fleshy hairs. M. braunsi Purcell. 2. “ Flagellum” similar, but much larger and more conspicuous. Head-plate strongly convex, the posterior surface rising vertically, the surfaces thickly covered with very short and numerous blackish brown cylindrical or pointed bristles. Four small recurved teeth in the single series of the upper jaw. Second abdominal sternite with 8-12 pairs of filiform fleshy hairs. M. globiceps Purcell. Annals of the Transvaal Museum 63 Genus Toreus Purcell. Toreus capensis Purcell, 1899. Annals S. Af. Mus. 1. p. 399, figs. 15-15 b. Originally described as a species of Ceroma, the genus Toreus being after- wards founded for its reception ( Annals S. Af. Mus. in. p. 9, fig. 7, 1903). The type came from the farm Bergvliet in the Cape Peninsula. According to the description, the most striking feature of the species is that presented by the fang of the upper jaw, which is long and slender and divided into two portions: the shorter basal portion is directed forwards: the longer distal portion, which is a continuation of the median half of the basal portion, is bent sharply outwards from the base, forming a right angle with its fellow of the other jaw. The lower jaw is also strongly curved upwards and outwards at the apex. The single series of teeth in the upper jaw is represented only by a low black crenular keel without any separate teeth. Text fig. 11. Ceroma pictulum Pocock. Mesial surface of upper jaw, with base of flagellum. Genus Ceroma Karsch. Ceroma sclateri Purcell, 1899. Annals S. Af. Mus. 1. p. 395, fig. 13. The type male came from Houwhoek, Caledon div. C.P. An immature specimen was also recorded from Ashton, Robertson div. It was recorded from Damaraland in Das Tierreich, but this was evidently an error, as is now admitted by Kraepelin. Ceroma inerme Purcell, 1899. Annals S. Af. Mus. 1. p. 398, fig. 14. The type male came from “ Walfish Bay.” Ceroma pallidum Pocock, 1900. Ann. Mag. Nat. Hist. 7, v. p. 305, fig. 7. The type male is from Garies in Little Namaqualand. Ceroma pictulum Pocock [PI. V, fig. 20 and Text fig. 1 1] , 1902. Ann. Mag. Nat. Hist. 7, x. p. 8, PI. II, fig. 2. 64 Annals of the Transvaal Museum The type male was taken at Teafontein near Grahamstown. The species described by me under the name of C. leppanae (. Records Albany Museum, hi. p. io, fig. 3) is probably referable to pictulum, although according to the donor of the specimen it came from Bulawayo. Mr S. Hirst has recently supplied me with notes on the type of pictulum as follows : flagellum very long and fine, reaching well beyond the hind margin of the head-plate, slightly expanded and truncate at the tip, the truncated edge being frayed out into short pointed processes, visible under the low power of a compound microscope: on the second and third legs, the patella has only a single very short spine at the. distal end, and the tibia has three dorsal spines. In the type of leppanae, there are no stridulatory ridges on the chelicerae: the upper surfaces of the chelicerae carry long stiffish bristles: the bristles protecting the base of the flagellum are all feathered : ocular tubercle elevated, being slightly but distinctly higher above the head-plate than in Solpuga, Daesia, Blossia or Chelypus: tarsus IV not spined: claws of tarsus I only just visible under a magnification of about 10 diameters. The compact fan of feather bristles, protecting the base of the flagellum, arises from a short outstanding projection of the surface of the chelicera, and is. presumably derived from the distal portion of the series of feather bristles that normally fringes the jaw in this family: nevertheless that series is represented in piciulum by a row of weakly developed feather bristles, more basally situated, and separated by an interval from the fan of stout bristles. The fan includes also one or two weak spines or simple bristles. As usual there is a row of simple bristles parallel to that of feathered ones. The shaft of the flagellum resembles that of Chelypus. It is furrowed along its length and seems to be essentially a folded membrane: basally the furrow passes into a small inflation on the mesial side. It is rotatable at the base, a fact which seems to have been overlooked by Pocock, as well as by myself when describing leppanae : possibly the flagellum of other species may be fixed, for no one has hitherto recorded a 'rotatable flagellum in this genus. Ceroma focki Kraepelin, 1914. Beit. z. Kenntnis d. Land- u. Susswasserfauna. Deutsch-Sudwestafrikas, p. 134, fig. 5. The type male is labelled Windhuk. Key to the S. African species of Ceroma. A. The single series of teeth in the upper jaw composed of four teeth in a continuous row, the third being small. Flagellum comparatively short, considerably expanded in the basal half but narrower towards, the apex. 1. Apex of flagellum filiform: the basal expanded portion protected on the mesial side by one strong spine and beneath it two bristles which are stout at the base and filiform at the^apex. C. pallidum Pocock. 2. Apical portion of flagellum stouter, and bent strongly backwards into a hook: on the mesial side, the flagellum is flanked by two strong spines one near the base and one near the apex, whilst on the outer side there are three stout spines near the basal expanded portion. C. focki Kraepelin. B. The single series of teeth in the upper jaw with only two teeth,, the basal one much the larger, the distal tooth considerably separated Annals of the Transvaal Museum 65 therefrom, being situated slightty in front of the anterior bend of the flagellum: flagellum long, reaching back well beyond the hind margin of the head-plate, not expanded into a pear-shaped enlargement at the base. Tibiae II and III with three short stout dorsal spines. C. pictulum Pocock. C. The single series of teeth in the upper jaw comprising two teeth, both rather large, not appreciably separated from each other. Flagellum long and slender. 1. Flagellum with a pear-shaped enlargement near the anterior bend, its apex reaching almost to the hind margin of the head-plate: tibiae II and III with thiee short stout spines dorsally. C. sclaieri Purcell. 2. Flagellum strongly compressed laterally at the anterior bend, but at the basal end it is scarcely enlarged, its apex reaching to the hind margin of the ocular tubercle: tibiae II and III with several strong bristles dorsally. C. inerme Purcell. Genus Lipophaga Purcell. Lipophaga trispinosa Purcell, 1903. Annals S. Af. Mus. in. p. 11, fig. 8. The type came from Stompneus and Soldatenpost at St Helena Bay, C.P. Lipophaga schultzei Kraepelin, 1908. Denh. med. nat. Gesell. Jena, xm. p. 280, figs. 9-10. This was originally described as the type of a distinct genus, Pseudoblossia Krpln., since recognised by its author as identical with Lipophaga. The types came from Kamaggas in Little Narnaqualand. Lipophaga michaelseni Kraepelin, 1914. Beit. z. Kennt. d. Land- und Siiss- wasserfauna Deutsch-Sudwestafrikas, Skorpiones und Solifugae, p. 132, fig. 4. The type came from Luderitzbucht . A female from Okahandj a was described at the same time, and referred with some doubt to this species. Key to the species of Lipophaga Purcell. 1. The single dental series of upper jaw with six teeth, the third and sixth being largest, the first of moderate size. Terminal fang of upper jaw much shorter than that of lower jaw. L. trispinosa Purcell. 2. Similar to trispinosa , but first tooth of upper jaw very much larger than the third, being in fact the largest of the series. L. michaelseni Kraepelin. 3. Terminal fang of upper jaw as long as that of the lower jaw. L. schultzei Kraepelin. In Kraepelin’s definition of the genus Pseudoblossia, the tarsus of the fourth leg is said to be at least six times as long as deep, and the first leg without tarsal claws: the chelicerae have stridulatory ridges reaching almost up to the middle of the inner surface, the male with about 16 such ridges. Purcell’s definition of Lipophaga includes : tarsus I with a pair of minute claws : chelicerae with a series of very short ridges at the anterior edge of the large smooth area. 5 66 Annals of the Transvaal Museum Kraepelin referred the genus to the Daesiinae, but Purcell included it in the Karschiinae. The species are of moderate size, the adult male measuring 15I-17 mm. in total length, and the female 20^—22. A series of female specimens belonging to this genus was taken recently by Dr R. Broom in the Campbell district: according to Dr Broom, the species is found over the dolomite area, extending from a point about 40 miles north of Campbell to 30 miles north-west of that village. In these examples the first tarsus is clawed: the tarsus of the fourth leg is about 5-6 times as long as deep, has no ungual appendages, and is rather densely clothed interiorly with stiff setae: tibia III with a row of three very strong dorsal spines, also three weaker ones in a row posterior thereto : patella III with one strong spine at the apex dorsally: fourth leg without spines: the chelicerae have about 15 comparatively short stridulatory ridges, as shewn in the adjoining figure: the single series of teeth in the upper jaw is unbroken, the first and second teeth being of moderate size, the third large, the fourth moderate, the fifth and sixth rather large ; coxa of palp with a long maxillary process : anterior margin of head-plate a little curved, upper surface with a well marked mesial groove and armed with long stiff setae and hairs like those on the chelicerae, with two Text fig. 12. Lipophaga sp. Female example frcm neighbourhood of Campbell, shewing: a, the stridulatory ridges on the inner surface of the chelicera; b, the jaws viewed from the outer side. prominent bristles in front of the ocular tubercle: the lateral portions of the head-plate are separated off from the mesial portion by a continuous suture as in Chelypus: in the interarticular membrane between the fourth and fifth sternites there is a distinct but minute median stigma : sides of abdomen only very sparsely hairy, the hairs being stiff and outstanding, with conspicuous pits arranged roughly in transverse rows. The largest example measures 28 mm. in total length (including the chelicerae), the head-plate being 5 mm. broad. The family Hexisopodidae was constituted by Mr R. I. Pocock (Ann. Mag. Nat. Hist. 6, xx. p. 250) for the reception of the remarkable genus Hexisopus, which at that time was very imperfectly known. The adult male had not been described and one of the two known species was based on a very young specimen, so that a character now known to be common Genus Hexisopus Karsch. Annals of the Transvaal Museum 67 to juveniles throughout the order, viz. the presence of only three malleoli on the hind limb, was represented as characteristic of Hexisopus and was even included in the definition of the family Hexisopodidae. Other more important characters were of course mentioned in the original definition and all later writers have agreed in recognising the family. The best account of the genus is that given by Dr Purcell (9) who sums up the principal points of distinction from the Solpugidae as follows: the absence of one joint of the trochanters in each leg of the three pos- terior pairs : the absence of claws in leg IV : the extreme shortness of the distal portion of leg IV in proportion to the basal part : the length of the distal segment of the claws in legs II and III : the minuteness of the third tarsal segments in these legs : the absence of subungual appendages in these legs: the presence of a genital papilla in the male: the shortness of the upper lobe of the rostrum. Prof. Kraepelin in Das Tierreich also relies mainly on the characters of the legs in distinguishing this family from the Solpugidae: the great length of the segment now termed the trochantin and the shortness of the femur of the fourth leg is cited as an important character, and indeed the difference from the other known S. African genera in this respect is very striking. Other characters mentioned by Kraepelin in his diagnosis of the family, in addition to some of those emphasized by Dr Purcell, are: — mandibles with anastomosing stridulatory ridges: mandible of male almost toothless: ocular tubercle conical, forwardly projecting, and very hairy anteriorly: characters of the male flagellum. In my opinion none of these characters can be regarded as equivalent in importance to those which distinguish the two main families of Solifugae, viz. the Galeodidae and the Solpugidae. The great reduction in length of the distal segments of the legs, the strong development of spines on these segments and the disappearance of the tarsal claws of the fourth leg are doubtlessly to be regarded as adaptation characters, in accordance with the burrowing habit : such differences between Hexisopus and a typical Solpugid can be compared with those which exist between our short limbed, stout bodied, burrowing frogs ( Rana delalandi Tschudi) and the long limbed, slender, grass frogs ( Rana fasciata) of the same genus, and indeed are not very much greater than those which separate the sedentary females of the genus Stasimopus (trap-door spider) from the free-living adult male of the same species. The supposed reduction in number of the trochanter segments is incorrect, being based on an error of homology. The number of tarsal segments, the absence of sub- ungual appendages from the tarsi and the absence of claws on legs IV are characters of no great importance in view of the wide variation that is now known to occur on the several legs of true Solpugids. The short- ness of the upper lobe of the rostrum is apparently a feature peculiar to Hexisopus, and is not found in the closely related genus Chelypus. The stridulatory area of the chelicerae, which in Chelypus is not ribbed but marked with fine and quite irregular furrows,. seems to be much more like that of a normal Solpugid in some species of Hexisopus where parallel stridulating ribs are present : in any case, however, the character is not of great value seeing that in the same species of Solpuga ribs may be 68 Annals of the Transvaal Museum fully developed in the female but quite absent in the male. The ocular characters vary very considerably in the several Hexisopodid species, and even in the same species between male and female. The dentition of the chelicerae is a most unsatisfactory character from a systematic ! point of view. The differences between the families Solpugidae and Galeodidae are apparently of a major- type, at any rate not connected with habitudinal differences. The Galeodidae have the stigmata of the second and third abdominal segments opening beneath the posterior edge of the sternites and protected by a pair of transversely elongated finely toothed plates the Serrulae: the stalks of the tarsal claws are hairy: the tarsus of the palp is freely articulated with the tibia. In the Solpugidae there are no serrulae protecting the abdominal stigmata (the modified hairs on the sternites in the Daesiinae may be homologous therewith), which are plainly visible on the posterior margins of the sternites : the stalks of the tarsal claws are not hairy : the tarsus of the palp is immovably attached to the tibia, or is only slightly movable thereon as in Daesia. In these characters, the Hexisopodidae agree entirely with the Solpugidae and I therefore propose to unite the two into one family, Solpugidae, where the Hexisopodid genera have the subsidiary rank of a subfamily, the Hexisopodinae. In respect to the form of the head-plate and of the first thoracic tergite, the Hexisopodid genera seem to shew considerable resemblance to the genus Rhagodes as figured by Mr H. M. Bernard in his paper on the ‘'Comparative Morphology of the Galeodidae” {Trans. Linn. Soc. Lon. 2nd Ser. Zoology, vol. vi. PI. 27, figs. 6 and 7). These genera also re- semble Rhagodes, more or less clearly, in the dentition of the chelicerae, in the shortness, of the legs, in the small size of the eyes, and in the rather large size of the anal segment. Assuming the accuracy of Mr Bernard’s figure, there is another interesting point in which Rhagodes differs from ordinary Solpugids and tends towards Hexisopus : the claw of the fourth leg is represented as smaller than that of the third leg, whereas in all genera found in S. Africa, except Hexisopus and Chelypus, the claw of the fourth leg is greater than that of any of the preceding legs. I may add that the Rhagodinae and Hexisopodinae will probably be found to occupy adjoining geographical areas, the genus Rhagodes extending southwards as far as Masailand, and the western genus Chelypus being known to me from North West Rhodesia. However, in the absence of actual material for an exact comparison between Rhagodes and the Hexisopid genera, it would be unwise as yet to lay too much stress on these points, which may prove to be explicable as the results of conver- gent evolution rather than of genetic affinity. At present, indeed, it is impossible to assert with any degree of confidence whether or not the Hexisopodid genera should be regarded as derivatives of the same immediate stock as gave rise to the other Solpugid genera in S. Africa. The Hexisopodinae are probably more nearly related to the Karschiinae than to either of the other S. African subfamilies. I have observed one character in Chelypus, and less distinctly in Annals of the Transvaal Museum 69 Hexisopus, that does not appear to be present in the better known S. African genera: this is the median tracheal stigma of the fourth abdominal sternite, which however is quite small. The same stigma occurs in Galeodes, but not in Rhagodes according to H. M. Bernard1. I have noticed it in female examples of the genus Lipophaga. Hexisopus lanatus C. L. Koch [PI. VIII, fig. 41], 1842. Purcell in Annals S. Af. Mus. 1. p. 384, figs. 1-7 b, also n. p. 221. Recorded by Dr Purcell from localities in Namaqualand, C.P., viz. Gras- mond between Steinkopf and Ramond's Drift, and from Vuurdood near Ramond’s Drift, also from Bladgrond in Gt Bushmanland. Kraepelin records it from Keetmanshoop and Kabus in Gt Namaland. Hexisopus crassus Purcell, 1899. Annals S. Af. Mus. 1. p. 387, figs. 8 and 8 a. This is founded on a single female example from Worcester. C.P. Hexisopus nigrolunatus Kraepelin, 1899. Das Tierreich, p. 15 1, fig. 117. Founded on an adult male from Damaraland. Hexisopus infuscatus Kraepelin, 1899. Das Tievreich, p. 151, fig. 118. The type is an adult male from Walfish Bay. The species does not seem to be very different from lanatus judging from the description. Hexisopus reticulatus Purcell, 1901. Annals S. Af. Mus. n. p. 222, fig. 9 a and b. The type is an adult male taken near the Orange River between Bysteek and the Great Falls at Aughrabies. Hexisopus fodiens Simon, 1887. Ann. Soc. Ent. France, 6. 7, p. 374, PI. VI, figs. 6 and 7. The type is an immature example from the Kalahari, and presumably the species is indeterminable. Dr Purcell has remarked on certain peculiarities (9) in Simon’s figure of this species, and I may add also that the distal portion of the palp is represented with quite unique features — having apparently a lobe on the tarsus, and the latter freely articulated with the tibia — which however are not referred to in Simon’s description. The figure is probably hopelessly inaccurate. Key to the species of the Genus Hexisopus . 1. From the anterior edge of the basal cup, the shaft of the flagellum passes forwards, upwards, backwards and finally downwards in a bold curve, terminating just behind the base of the lower jaw, not far from the basal cup : the shaft is more or less cylindrical and tapers gradually to a slender apex. H. lanatus C. L. Koch. 1 According to Sorensen’s account, the median spiracle would seem to be normally present in Solifugae. Hansen recorded it in the genus Rhagodes. 5—3 70 Annals of the Transvaal Museum 2. The shaft of the flagellum is of more uniform thickness and does not taper gradually to a slender apex, the distal end being strongly compressed, blade-like, and slightly dilated: the shaft moreover is not curved throughout its length, the upper portion being directed back- wards in a straight line for a considerable distance. Lower jaw long, slender, strongly curved, with a small tooth behind the middle. H. reticulatus Purcell. 3. Towards the apex of the flagellum, the shaft presents for a short distance a small membranous expansion along the lower side: this terminates quite abruptly near to the distal end which is drawn out to a fine point. The lower jaw has a fairly large tooth with a horizontal upper edge, the distal end of the tooth being nearer to the apex of the jaw than to its base. Upper jaw with two small teeth near to the point of origin of the flagellum. H. nigrolunatus Kraepelin. 4. Similar to lanatus, but tibia of palp with short cylinder bristles over the whole extent of the upper surface, the cylinder bristles of the lower surface being short distally, but those near the base of the segment gradually become longer, exceeding twice the length of the distal bristles. (In lanatus the tibia of the palp, dorsally and ventrally, is said to carry long hairs, rather long bristles and between them some short isolated cylinder bristles.) H. infuscatus Kraepelin. Genus Chelypus Purcell. Chelypus barberi Purcell, 1901. Annals S. Af. Mus. n. p. 224, fig. 10. The type is a male from a locality in Bechuanaland about 100 miles south of the junction of the Moshowing with the Molopo Rivers. The Transvaal Museum has a specimen labelled Rietfontein, Gordonia, which seems referable to this species. In this specimen, the chelicerae are not densely covered with minute pointed spinules on the upper surface, but have short subspiniform setae and some very minute spinules. The cup-like basal enlargement of the flagellum narrows into a rounded solid head which fits closely into a hemi- spherical depression on the surface of the chelicera. Chelypus lennoxae Hewitt, 1912. Records Albany Mus. n. p. 312. The type came from Upington. This species is remarkable in its dark colouration, the hairs of the abdomen superiorly being brown or dark brown, and the sides quite deeply infuscated : the fourth femur is rather long when compared with the trochantin: the chelicera has comparatively coarse, black, pointed spinules on the outer portion of the upper surface and on the upper portion of the outer surface, whilst a prominent oblique patch of strong pointed spinules occurs on the mesial surface near to the base of the fang superiorly: the furrows of the stridulatory area are very long, and in general longitudinally arranged: a distinct ocular tubercle is not present, but the area on the mesial side of each eye is slightly raised: the abdominal tergites are well defined. Chelypus hirsti Hewitt [PL VIII, fig. 45 and Text-fig. 12], 1915. Annals Natal Mus. in. p. 323, figs. 8 and 9. The types were taken at Rietfontein in Gordonia. Annals of the Transvaal Museum 7i The following notes may be added to the original description: eyes about seven or eight diameters apart, being situated on a more or less distinct ocular tubercle which is quite twice as broad as long and almost subdivided by a median depression into two tubercles : this ocular region proj ects forwards a little in front of the more lateral portions of the head-plate but is not nearly so forwardly produced as in the males of Hexisopus, where the eyes are much nearer together according to Purcell’s accounts. Inner surface of chelicerae without feathered bristles, and with no row of spines nor even stiff bristles on the upper jaw: stridulatory area rather large, with a reticulation of furrows. Text fig. 13. Chelypus hirsti Hewitt, a, Head-plate and thoracic tergites seen from above, b, Head-plate seen from the side, c, Third leg viewed from its upper side (this leg is considerably twisted), d, Distal segments of same, more enlarged, in ventral view, e, Second leg and / fourth leg. Coxa of palps with a fairly prominent maxillary lobe : this is not so long as that of Solpuga, but is longer and more conspicuous than in Daesia. The palp has very stout spines on its surfaces, four on the tarsus, about 11 on the tibia, and several on the patella. Claws of second and third legs of two seg- ments, but the basal segment is almost obsolete, being about one-twelfth as long as the distal segment: according to Dr Purcell, this character may vary considerably amongst the individuals of the same species {Hexisopus lanatus). The anal slit is large, and the whole segment moderately large. A small median spiracle on the fourth abdominal sternite, which is also traceable in 72 Annals of the Transvaal Museum the male of Hexisopus lanatus but only indistinctly in the female: thoracic spiracles very large. There is a general absence of bristles or stiff hairs on the surfaces of body and appendages, but numerous long soft and low-lying hairs occur. Key to the S. African species of Genus Chelypus . 1. Shaft of flagellum directed forwards, upwards, backwards, and downwards in a bold curve, the apex bifid, with a short tooth-like medial branch and a longer lateral branch. Dorsal surface of upper jaw with two very large black tubercles distally. Tibia of fourth leg longer than the tarsus, and broader than the patella, but not lobed : tarsus of fourth leg broader than long. Claws of second and third leg composed almost entirely of the distal segment, the proximal segment (claw peduncle) being extremely short and indistinctly marked off. C. barberi Purcell. 2. Similar to barberi but flagellum not bifid at the apex but tapering to a point. C. hirsti Hewitt. 3. Dorsal surface of upper jaw with only one large tubercle. Tibia of fourth leg with a large rounded lobe projecting externally and densely covered with short spinules. Tarsus of fourth leg longer than broad. Distal segment (claw proper) of tarsal claw of third leg less than half the total length of the claw and distinct^ marked off from the proximal segment (claw peduncle). C. lennoxae Hewitt. Annals of the Transvaal Museum 73 ADDENDUM Solpuga niassa Karsch var. nov. kafulica [Text fig. 14]. What seems to be a distinct variety of this species is known to me from Kafula Futa, N. Rhodesia (Rev. C. M. Doke). The male presents considerable features in common with schonlandi, and, I think, will probably prove to be completely connected therewith by intermediate forms. It differs from schonlandi as follows: the flagellum is more slender and a little longer, the apex reaching back beyond the hind border of the basal enlargement, the recurrent portion of the shaft tapering from its commencement, rather rapidly so near the anterior bend, becoming somewhat twisted outwards distally, the apex rather bluntly pointed and with no distinct projecting core: the terminal fang of the upper jaw has the inner dorsal tooth almost obsolete: the lower border of the terminal fang well curved. Only one intermediate tooth in the single series of the upper jaw. Basal enlargement of flagellum markedly elevated as in schonlandi. Text fig. 14. Solpuga niassa kafulica sub sp. nov. Left upper jaw of male viewed from mesial side. In the female, there are indications of a second intermediate tooth in the upper jaw: this, which is actually the third tooth, is much smaller than the fourth, which again is considerably less than either the second or fifth. Measurements: breadth of head-plate M. 7, F. 13: length of patella of palp M. 12, F. 14: of tibia and tarsus of palp M. 12-3, F. 14-8: of patella of fourth leg M. 10-5, F. 12-25. According to Kraepelin/s figure of niassa in Das Tievreich, the upper jaw of the male has two well developed intermediate teeth, and the flagellum has peculiar serrations on its shaft, which are not in any degree represented in the specimen from Kafula Futa. The new variety is related also to 5. rhodesiana Hirst ( Manchester Memoirs, lvi. 1911, No. 2, p. 8) which has two intermediate teeth in the upper jaw, whilst the flagellum is sharply pointed at the apex and distinctly shorter than that of kafulica. 74 Annals of the Transvaal Museum Fig. i. Fig. 2. Fig. 3- Fig. 4. Fig. 5- Fig. 6. Fig. 7- Fig. 8. Fig. 9- Fig. 10. Fig. 11. Fig. 12. Fig. 13. Fig. 14- Fig. 15- Fig. 16. Fig. 17. Fig. 18. Fig. 19. Fig. 20. Fig. 21. Fig. 22. Fig. 23. Fig. 24. Fig. 25. Fig. 26. Fig. 27. Fig. 28. Fig. 29. Fig. 30. Fig. 31- Fig. 32. Fig. 33- EXPLANATION OF PLATES PLATE I. Solpuga Venator Poc. Adult female from Keetmanshoop. PLATE II. Solpuga chelicovnis Licht, var. nov. pubescens. Adult male from De Aar. 5. chelicovnis Licht, var. nov. rufescens. Adult male from Longhope. 5. derbiana Poc. Adult female from Grahamstown. PLATE III. Hemiblossia O’neili Purcell. Adult female from Alicedale: enlarged. Solpuga sericea Poc. ? sp. (possibly celeripes Hirst). Female from Salisbury. S’, lineata Koch. Male from Alicedale. 5. chelicovnis Licht, var. nov. rufescens. Female from Dirkskraal. 5. Venator Poc. Chelicera of adult female from Thorn Kloof near Grahams- town: enlarged. 5. Venator Poc. Abdomen of adult female in ventral view. S. chelicovnis Licht. Abdomen of adult female in ventral view. PLATE IV. Solpuga Venator Poc. Chelicera of adult male from Alicedale : enlarged. S. strepsiceros Kraep. Chelicera of adult male from Barberton: enlarged. S', bechuanica Hewitt. Chelicera of adult male from Serowe: enlarged. S. schlechtevi Pure. Chelicera of adult male from Keimoes : enlarged. S. hostilis White. Ventral surface of trunk of adult male from Pretoria. S. hostilis White. Ventral surface of trunk of adult female from Pretoria. PLATE V. Solpuga spiralicornis Purcell. Chelicera of adult male from Griffin Mine, Leydsdorp: enlarged. S. ovnithovhyncha Hewitt. Chelicera of adult male from the neighbourhood of Kraikluft: enlarged. Ceroma pictulum Poc. Chelicera of adult male described as C. leppanae mihi: enlarged. Solpuga globicornis Kraep. Chelicera of adult male from Pretoria dist. : enlarged. S. suffusca Hewitt. Chelicera of adult male: enlarged. S. mavaisi Hewitt. Chelicera of adult male from Stellenbosch: enlarged. S. lateralis Koch. Chelicera of male from Bussacks : enlarged. PLATE VI. Zeriassa puvcelli Hewitt. Chelicera of female specimen from Newington: enlarged. Solpuga fusca Koch. Chelicera of female specimen from Capetown: en- larged. Daesia lineata (Pocock). Chelicera of adult male from De Aar: enlarged. D. lineata (Poc.). Female example from De Aar. Solpuga hostilis White. Chelicera of male example from Doornkop : enlarged. S. mavshalli Poc. Chelicera of adult male from Salisbury: enlarged. Daesia lineata (Poc.). Male example from De Aar: enlarged. Solpuga celeripes Hirst. Chelicera of adult male: locality unknown. S. lineata Koch. Chelicera of adult male from Alicedale: enlarged. Annals of the Transvaal Museum 75 PLATE VII. Solpuga alcicornis Kraep. Chelicera of adult male from Kuruman: enlarged. S. tubicen Kraep. Chelicera of adult male from locality unknown: enlarged. Zeriassa purcelli Hewitt. Chelicera of male from Newington: enlarged. 5. lethalis rectus var. nov. Chelicera of male from Windhuk: enlarged. 5. monteiroi Poc. Chelicera of male from Rustenburg dist. Blossia unguicornis Purcell. Adult female from Alicedale: enlarged. B. unguicornis Purcell. Adult male from Alicedale: enlarged. PLATE VIII. Hexisopus lanatus Koch. Male example in ventral view: enlarged. Hemiblossia O'neili Purcell. Fourth leg of female example from Alicedale: enlarged. Daesia lineata (Poc.). Ventral view of anterior portion of abdomen of male, shewing first three sternites and basal segments of fourth leg: enlarged. Blossia unguicornis Purcell. Ventral view of abdomen of male: enlarged. Chelypus hirsti Hewitt. Chelicera of male from Rietfontein, Gordonia : enlarged. Blossia unguicornis Purcell. Fourth leg of male from Alicedale: enlarged. LIST OF RECENT LITERATURE ON THE S. AFRICAN SOLIFUGAE (1) Kraepelin, K. Palpigradi und Solifugae, in Das Tierreich, Lief. 12, 1901. (This work includes complete lists of references to genera and species then known , as well as short descriptions and figures.) (2) Skorpione und Solifugae, in L. Schultze, Forschungsreise im westl. und Zentr. Sudafrika • 1903-05. Jena, 1908, pp. 247-282. (3) Skorpiones und Solifugae, in Beitr. zur Kenntnis der Land- und Susswasser- fauna Deutsch-Siidwestafrikas . Hamburg, 1914, pp. 107-136. (4) Neue Beitrage zur Systematik der Gliederspinnen, in Mit. aus d. Nat. Museum Hamburg, xxvm. pp. 59-107. 1911. (5) Pocock, R. I. On the genera and species of Tropical African Solifugae, in Ann. Mag. Nat. Hist, 6, xx. pp. 249-272. 1897. (This paper summarises the differential characters of the African species and genera as known at that time.) (6) On some new or little known Thelyphonidae and Solifugae, in Ann. Mag. Nat. Hist. 7, v. pp. 294-306. 1900. (7) Some new Arachnida from Cape Colony, in Ann. Mag. Nat. Hist. 7, vi. pp. 316- 333. 1900. (8) Descriptions of some new species of African Solifugae and Araneae, in Ann. Mag. Nat. Hist. 7, x. pp. 6-27. 1902. (9) Purcell, W. F. New and little known South African Solifugae in the collection of the South African Museum, in Ann. S. Af. Mus. 1. pp. 381-432. 1899. (10) On some South African Arachnida belonging to the orders Scorpiones, Pedipalpi and Solifugae, in Ann. S. Af. Mus. 11. pp. 137-224. 1901. ( 1 1 ) Description of new genera and species of South African Solpugidae, in Ann. S. Af. Mus. ill. pp. 1-12. 1903. (12) New Arachnida collected by Mr S. C. Cronwright-Schreiner at Hanover, Cape Colony, in Ann. S. Af. Mus. hi. pp. 13-40. 1903. (13) On the Scorpions, Solifugae and a Trapdoor Spider collected by Rev. Henri A. Junod at Shilouvane near Leydsdorp in the Transvaal, in Novitates Zoologicae, x. pp. 303-306. 1903. Fig. 34 Fig. 35 Fig. 36 Fig. 37 Fig. 38 Fig. 39 Fig. 4° Fig. 41. Fig. 42. Fig. 43- Fig. 44. Fig. 45- Fig. 46- 7 6 Annals of the Transvaal Museum (14) Hewitt, J. Description of a new species of Chelypus, in Records Albany Museum, 11. pp. 312, 313. 1912. (x5) Descriptions of new species of Arachnida from Cape Colony, in Records Albany Museum, 11. pp. 462-481. 1913. (16) Records and descriptions of the Arachnida collected by the Percy Sladen Memorial Expedition to Great Namaqualand, 1912-1913, in Annals Transvaal Mus. iv. pp. 146-159. 1914. ( 1 7) Records of species of Solifugae in the collection of the Transvaal Museum and descriptions of several new species of the family Solpugidae, in Annals Transvaal Mus. tv. pp. 160-167. I9I4- (18) Descriptions of new Arachnida from S. Africa, in Records Albany Mus. hi. pp. 1-37. 1914. (19) Descriptions of new South African Arachnida, in Records Albany Mus. hi. pp. 70-106. 1915. (20) — — New South African Arachnida, in Annals Natal Mus. in. pp. 289-327. 1915- (21) Descriptions of several species of Arachnida in the collection of the Durban Museum, in Annals Durban Mus. 1. pp. 217-227. 1916. (22) Descriptions of new South African Arachnida, in Annals Natal Mus. 111. pp. 687-711. 1917. (23) Description of new South African Araneae and Solifugae, in Annals Transvaal Mus. vi. pp. 63-106. 1918. (24) Hirst, A. S. On a collection of Arachnida and Chilopoda made by Mr S. A. Neave in Rhodesia, North of the Zambezi, in Manchester Memoirs, lvi. No. 2, pp. 1-11. 1911. (25) Descriptions of new Arachnids of the Orders Solifugae and Pedipalpi, in Ann. Mag. Nat. Hist. 8, ix. pp. 229-237. 1912. (26) On a new species of Solpuga from Zululand, in Annals Durban Museum, 1. pp. 228-229. 1916. MORPHOLOGICAL. (27) Bernard, H. M. On the Comparative Morphology of the -Galeodidae, in Trans. Linn. Soc. London, 2nd ser. Zoology, vi. pp. 305-417. 1896. (28) Sorensen, W. Recherches sur l’anatomie, exterieure et interieure, des Solifuges, in Oversigt o. d. K. d. Videnskabernes Selskabs F orhandlinger , No. 3, pp. 131- 215. 1914. Kjobenhavn. Annals Transvaal Mus. Vol. VII , PL i Plate II Diurnal Solifugae Annals Transvaal Mus. Vol. VII, Pt. i Plate III Annals Transvaal Mus. Vol. VII, Pt. i Plate IV Annals Transvaal Mus. Vol. VII, Pt. i Plate V 23 24 Annals Transvaal Mus. Vol. VII, Pt. i Plate VI Annals Transvaal Mus. Vol. VII, Pt. i Plate VII Annals Transvaal Mus. Vol. VII, Pt. i Plate VIII « 45 ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM VOLUME VII ■LjjU Muse^ PART 2 containing Contributions to the Knowledge of the Reptiles of the Karroo Formation. By Dr E. C. N. van Hoepen, M.I. 5. A New Dinosaur from the Stormberg Beds. (With 2 plates and 6 text-figures.) 6. Further Dinosaurian Material in the Transvaal Museum. (With 13 plates and 27 text- figures.) Description of some Cretaceous Ammonites from Pondoland. By Dr E. C. N. van Hoepen, M.I. (With 3 plates.) Issued March 10 th, 1920 PRINTED AT THE UNIVERSITY PRESS CAMBRIDGE, ENGLAND 1920 0 ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM Vol. 7 Part 2 CONTRIBUTIONS TO THE KNOWLEDGE OF THE REPTILES OF THE KARROO FORMATION In November of 1915 the Transvaal Museum bought the remains of a small Dinosaur from a resident of Ficksburg. The specimen had been discovered in a building-stone quarry near Roosendal, O.F.S. This is most probably the same quarry in which the small Pseudosuchian Pedeticosaurus Leviseuri has been found. During the usual quarrying work a stone was split and found to contain the remains of some animal. The little village of Roosendal was very interested in the find and in due course photographs of the fossil were sent to friends. One of these photographs reached my friend Mr A. W. Putterill of Harrismith, who sent it on to me with the necessary information. Through this report the Transvaal Museum eventually came into possession of one of its best specimens. I have therefore great pleasure in thanking Mr Putterill for the service he has rendered to our institution in particular and to science in general. It follows, from the above, that the remains are on two slabs of rock. They consist of a nearly complete skeleton. The skull is lost or at least not visible, for it is not impossible that it is present in a more or less fragmentary con- dition under some of the other bones. Two jaw portions with some teeth are exposed. The neck vertebrae have been crushed sideways into a heap. All the dorsal vertebrae are more or less articulated as also a fair number of caudals. The front legs are incomplete, but the hinder extremities are very well preserved. The shoulder girdle is only represented by two coracoids and the proximal ends of the scapulae, while the pelvis shows all its constituents of which some are in very good condition. All the bones, however, have been much flattened. By Dr E. C. N. van Hoepen, M.I. 5. A NEW DINOSAUR FROM THE STORMBERG With two plates and six text-figures 6 78 Annals of the Transvaal Museum Skull. A large part of a left maxillary and the front end of probably the left ramus of the lower jaw are present. The length of the lower maxillary border as pre- served is about 29 mm. It bears six teeth or remnants of teeth and in front of the foremost one is the crown-tip of another one, which was probably also on the maxillary. Behind these six the rock only shows a brown blur and nothing definite can be made out. The hinder end of the alveolar border is only indicated. The front end of the maxillary and the greater part of the front border of its nasal process are covered by the coracoid. The anterior lower corner of the antorbital vacuity is rounded and wide-angled. A portion of the inner wing of the nasal process is still present. The lower jaw portion is very much damaged and sutures cannot be made out. It shows three teeth and fragments of four others. Its lengthis about 2 5 mm. It seems as if the maxillary teeth slightly increase in size forwards. The four preserved tips of the dentary teeth occur at the same level and these four teeth were therefore probably of equal length. The roots of the two foremost ones have been partially laid open and a small replacing tooth is seen in front of the first one. Another young tooth, which is still below the edge of the sup- posed dentary, was laid bare through removal of part of the bone and the root of the fourth tooth present. The crowns of the teeth are broad and flat and have the edges serrated. They are not pointed but rounded at the top and on one side at least there is a low broad ridge, which runs from the base along the middle to the upper edge. The denticles of the serrated edge have a blunt point, and the notches are angular. This angle appears to be very large, in some cases where the notch is clearly visible it shows an angle of some 6o°. In young teeth, however, the condition seems to be different. The young tooth below the root of the fourth one shows a portion of its edge with four small denticles. In a lateral view these denticles are long and cylindrical, and before being well cleaned of matrix the notches seemed to have the shape of a long stretched U. After the matrix had been carefully removed, however, it ap- peared that the denticles are united to such an extent that only their tips are free. The actual notch is slight and nearly rectangular, but continues on to the tooth surface as a parallel concavity between the denticles for about three times its real depth. Length of the crown of the best preserved maxillary tooth 5 mm., breadth 3*5 mm. In this tooth three denticles occupy the space of 1 mm. The crown of the young tooth has a length of 3*5 mm. and its four denticles stand on 1*3 mm. Cervical Vertebrae. All the cervical vertebrae have been crushed sideways into a heap. They have been very much flattened and broken and only three are sufficiently free from the others to be of any use. Two of these are very long and their place seems to be in the hinder region of the neck. The length of the centra of both is 31 mm. The ventral surface of one centrum is exposed and the breadth of Fig. 1. Portion of maxillary and dentary with teeth as preserved. Nat. size. Annals of the Transvaal Museum 79 the bone is 9 mm. This ventral surface is deeply concave. The breadth of the hinder articulation surface is 14 mm.; that of the front end is smaller. Nothing is visible of diapophyses or parapophyses. The praezygapophyses are short, pointed and wide apart. The postzygapophyses are longer. The height of the front articulation surface- of the vertebra, where all these dimen- sions are taken from, is 10 mm. The third vertebra is very much shorter and is the only neck- vertebra showing a dorsal spine. The height of this vertebra, with its spine, measured in the middle of its length is 19 mm. The length of the spine ( i.e . from the anterior to the posterior end) is 18 mm. The height of the hinder articulation surface is 10 mm. This seems to be one of the fore- most vertebrae of the neck. Dorsal Vertebrae. The total number of dorsal vertebrae is 14. The vertebra which I take to be the first dorsal is largely covered by the coracoids. It is not impossible, although I think it very improbable, that there is another dorsal vertebra in front of this one. The first seven vertebrae are still articulating, but the eighth is displaced. The ninth to the fourteenth are also more or less in juxtaposition. They all lie on their left side and through the splitting of the stone all the vertebrae have been broken and greater or smaller portions of each are still in both slabs. Therefore, generally speaking, no detail of the outer surface of the vertebrae is visible. Their principal measurements are Dorsal vertebra No. 1 Length of the centrum Height of front articulation surface Height of hinder articulation surface Total height Length of dorsal spine 3 4 5 6 7 8 9 10 11 12 13 14 20 21 23 24 24 24 26 28 29 15 15 16 17 15 15 15 15 17 19 36 37 37 38 39 39 37 40 16 19 22 22 22 21 From these figures and from the plates it will be seen that the vertebrae do not differ much in size. The length of the centra seems to increase slightly and gradually towards the pelvis. The dorsal spines are thin, long and low. Their front upper corner is rounded and their upper hinder corner more pointed and overhanging. Their upper border is evenly rounded, and its general direc- tion diverges slightly backwards with the axis of the centrum. The prae- zygapophyses project far beyond the plane of the anterior articulation surface of the centrum. The postzygapophyses are shorter, but still project well be- yond the plane of the posterior articulation surface. There is a round opening between the praezygapophysis and the anterior vertebra, which apparently served as an exit for a spinal nerve. The vertebrae are amphicoelous, but the only articulation surface visible is the rather deep depression on the hinder end of the first vertebra. The neural arch is attached to the centrum by a suture, which is plainly visible in the third to the sixth, the eighth to the tenth and the twelfth vertebra. This suture is seen on the outer surface of the vertebra as a zig-zag line, the general direction of which is parallel to the axis of the centrum. The amplitude of the undulations of this line seems to lie within certain limits, which are reached by many individual waves. The connection seems to have been a loose one and could only have prevented the neural arch from moving in a direction parallel to the axis of the centrum. Apparently centra and neural arches became easily disconnected, for this is more or less the case in five vertebrae. Ribs are present near all the vertebrae except the fourteenth and it is very probable that this vertebra did not bear ribs. The animal would then have 6—2 8o Annals of the Transvaal Museum a single lumbar vertebra. The anterior ribs are long and slender, while the posterior ones are short. In some of the ribs the proximal end is complete and is seen to be double-headed. Sacral Vertebrae. There is not much visible of the sacral vertebrae. The second shows its ventral surface, while the third shows its left side and partly its ventral surface. The posterior end of the first sacral protrudes from under the proximal end of the right ischium. The length of the centrum of the third sacral is 15 mm., while that of the second is 26 mm. The anterior corners of the ventral surface of the second sacral are truncated, probably by the articulation faces for the sacral ribs. These truncations reach backwards till past the middle of the ventral surface. A similar truncation is visible at the left anterior corner of the surface shown by the third sacral. There is a large brown patch with remains of bone to the right of the second sacral. I take this to be the second sacral rib of the right side. It lies at a distance of about 1 cm. from the vertebra. As preserved, the proximal end of this rib is very narrow and the distal end very broad. The anterior edge shows the original margin of the bone, 'which is very concave. None of the other margins have been preserved, except the anterior portion of the distal margin. The anterior portion of the distal end is very thin, and its outer margin is convex ; its upper surface is concave. The upper surface of the posterior portion of the distal end seems to have been convex and the whole distal end therefore shows a wavy appearance. The breadth of the distal end as preserved is 31 mm. and that of the proximal end 14 mm. The length of the rib as preserved is 20 mm. Behind the second there are remains of the third sacral rib. This is accompanied by its fellow on the other side of the third sacral vertebra. Both ribs lie at a distance of about 1 cm. from the vertebra. They are very badly preserved, but the contour of the distal end of the right rib seems to be complete on the slab figured on Plate IX. The anterior and posterior margins of the ribs are concave and the distal end is broad. Its outer margin is convex. The third sacral rib is very much narrower than the second. The length of the rib on the right is 21 mm., the breadth of its distal end 14 mm. and that of its proximal end 7 mm. The third sacral vertebra has been identified as such through analogy with other Anchisauridae and Plateosauridae . The specimen itself does not show any character which would class it immediately as sacral; on the contrary, its size and the shape and size of its sacral ribs agree so closely with the corresponding properties of the first caudal vertebrae, that without the know- ledge of other Theropoda one would without doubt regard it as the first caudal. Moreover, there is some doubt with regard to the second sacral. There is a slight suggestion in its shape that it represents two coalesced vertebrae; if these are the first and second sacrals, then it is difficult to conclude what the small bone may be which protrudes from under the right ischium. Its con- vexity and its convex free border strongly suggest the hinder end of a lower surface of a vertebra. Could it perhaps be a fifteenth dorsal? I doubt it and therefore I have regarded it as the first sacral. However, what is here called second sacral may be the coalesced second and third. But this is very im- probable, because of the first sacral not being coalesced. Moreover, the second sacral rib seems to form an unbroken series with what has been styled the third sacral rib and the first three transverse processes; the actual third sacral rib would then be missing. I do not see any reason which forces one to this conclusion, and therefore I have described these bones as above. Still it is Annals of the Transvaal Museum 8i very probable that the third sacral rib was too short to reach the ileum, not- withstanding normal backward convergence of the ilea. However, this may be explained in the same way as certain differences in Theropod sacra by v. Huene (31, p. 281). As appears from many loose sutural connections, our animal is a young individual, and has not yet reached the stage of life in which it possesses a true third sacral vertebra. Caudal Vertebrae. Of the caudal vertebrae only twelve have been preserved. The first two caudals are present. The first caudal lies immediately behind the last sacral and the second caudal behind the first. Then follows a large gap. After this comes a series of eight vertebrae. Two more vertebrae can be recognised in a jumbled up mass at the end of this series. The centrum of the first caudal has a length of 15 mm. That of the second cannot be accurately measured, because it has split obliquely. It is also diffi- cult to measure the height of the centra, but the dimension seems to be slightly less or equal to the length. The lower portion of the hinder articulation surface of both centra has been obliquely truncated to form an articulation surface for the haemapophysis. In the second caudal it is clearly visible that the lower portion of the -front articulation surface bends slightly backwards. The transverse processes of the right side have both been disjointed and their impressions lie at a distance of about 2-5 cm. from their respective vertebrae. They lie in a row with the second and third sacral ribs. The process of the first caudal has a length of 18 mm. and a breadth of 10 mm. These measure- ments of that of the second caudal are respectively 16 mm. and 10 mm. Both processes show great resemblance to the third sacral rib. The series of eight vertebrae is from the middle region of the tail. The centra are very long and low and they all show their lower surface and that of their right transverse process. The dorsal spines of all these vertebrae are embedded in the rock and invisible. The centra are so much flattened, that it is impossible to give any detail. The articulation surfaces of the centra project beyond their sides and lower surface. The transverse processes are remarkable for their broad attachment to the centrum. In the first vertebrae of the series they are practically attached to the whole length of the centrum, while in the last the attachment is still about two-thirds of this length. The distal three-fifths of the process has parallel sides, but those of the proximal end diverge towards the centrum. The process was apparently supported by lateral ridges. There are indications that the front lateral ridge of the process was in touch, in all the vertebrae of the series, with the praezygapophysis. Haemapophyses are attached to all these vertebrae and still in situ. Measurements from the first vertebra in the series to the last : Length of centrum 17, 17, 18, 18, 18, 18, 19, 19 mm. Length of transverse process 18, 19, 18, 15, 15, 15, — , 12 mm. Length of haemapophyses 40, 37, 31, 31 mm. Breadth of their distal ends 7, 7, 7, 7 mm. In front of the first vertebra of the series lies another haemapophysis, which through its position gives the impression of having belonged to the first pre- ceding vertebra. This haemapophysis has a length of 42 mm. Four or five other haemapophyses are scattered about, one of which is perfect. This has a length of 45 mm. The distal end of all these haemapophyses is broad and very thin. The proximal end, distal to the foramen, is narrower and thicker. The foramen of the longest haemapophysis has a height of 7 mm. and a breadth 1 These numbers refer to the list of literature at the end of the article. 82 Annals of the Transvaal Museum of 5 mm. The articulation surface of this bone has a breadth of io mm. The scattered haemapophyses belong to the missing vertebrae between the second caudal and the series of eight. It seems as if the vertebrae in the mass near the end of the series belong to the proximal portion of the tail, but no re- markable features are visible. Shoulder Girdle. The scapulae are completely embedded in the rock. The right scapula only shows an indeterminable edge and the left only what are probably its articu- lation surfaces. One of these surfaces has a more or less triangular contour and this is probably the articulation surface for the humerus. The other is for the contact with the coracoid. Both surfaces are concave and they meet at an obtuse angle. Both coracoids are present. Half of the left coracoid, however, is hidden by the one of the right side. Both bones only show their lateral surface, and therefore nothing can be made out of their thickness or their marginal surfaces. The contour of the right coracoid is complete. The scapular border of the cora- coid is not straight, for slight- ly above its junction with the glenoidal border it is deeply indented. This in- dentation has a length of 12 mm. and a maximum depth of 3 mm. Further upwards the contour line pjg 2. Aristosauruserectus. Right and left coracoid bends slightly forwards. The and proximal surface of left scapula. The border upper border is not well pre- behind the supracoracoidal foramen in the left served but seems to be hori- coracoid is not clearly visible. About nat. size, zontal ; it would then make a nearly right angle with the front border. The upper portion of the front border is straight; below the level of the middle of the bone it curves backwards; the lower front border of the bone is straight. The glenoidal border is fairly straight and meets the lower portion of the front border practically at right angles. The surface of this coracoid is fairly flat and one would scarcely take notice of some slight undulations if the left cora- coid were not present. The visible portion of that bone is not flat for it shows a rather deep concavity on its outer surface, running from the hinder end of its glenoidal border upwards. The supracoracoidal foramen is situated in this concavity, which continues past this opening. The lower corner of both inner and outer surface of this coracoid bulges outwards. Both peculiarities are also seen in the right coracoid, but in a much less degree. As the bones are now reduced to the thinness of paper, it is not impossible that the described un- dulations are due to original differences in the thickness of the bone. However, it must be remembered, that the bulge in these coracoids is in practically the same position as the tuberositas coracobrachialis described by v. Huene on the coracoid of Pachysaurus magnus (3, p. 149). The upper and front portion of the bone is much thinner than the lower and hinder portion. The greatest height of the right coracoid is 43 mm. and its greatest breadth 32 mm. Both coracoids have a foramen supracoracoideum. It is fairly large and situated close to the middle of the scapular border. Its Annals of the Transvaal Museum 83 diameter is about 3 mm. and the distance from the scapular border is between 2 and 3 mm. Humerus. Both humeri are present. Their proximal ends are fairly well preserved, but their distal ends are poor. The very broad proximal end is hollow in front and convex behind. Its upper border is convex and its upper-lateral border slightly so. There is no distinct demarcation between the upper end of the processus lateralis and the upper border of the bone. The lateral border of the humerus is strongly concave immediately below the processus lateralis. The medial border of the proximal end is concave throughout. There are indications that the caput humeri was not situated on the extreme medial end of the upper border. The breadth of the proximal end, measured from the lower corner of the processus lateralis to the upper medial end, is 45 mm. in the left and 52 mm. in the right humerus. The shaft is very narrow; at its narrowest place it measures 9 mm. in the left humerus. All that can be said of the distal end is, that the condyli project forwards and do not make themselves apparent on the hinder surface. The length of the humerus differs considerably in the two specimens. The right humerus measures 93 mm. in length and the left only 80 mm. The length of the left humerus is most probably reduced by crushing, as it stands nearly vertically on the general plane of the remains. The lower end of the processus lateralis lies 53 mm. above the distal end in the left humerus and 58 mm. in the right one. Ulna and Radius. There is not much left of these two bones. Only those of the right side have been preserved and these are really nothing more than impressions. The proximal end of the ulna is broad. The line marking its articulation surface is not perpendicular to the axis of the bone, and therefore, what is most probably the hinder portion of this end, is higher than the front portion. The breadth of this end of the impression is 18 mm. (this would in reality be the thickness of the bone). Distally this dimension is gradually reduced towards the middle of the shaft, which has a breadth of 9 mm. Then the breadth increases again towards the distal end, which may have been as broad as the proximal end. The total length of the bone is 64 mm. The impression of the radius has a proximal breadth of 10 mm. and a distal breadth of 14 mm., while the shaft has a breadth of 8 mm. The total length of the bone is 58 mm. The line marking the proximal end of the im- pression seems to be concave, so that probably the proximal articulation surface was hollow, at least near the side which made the impression. Carpus. The remains of probably four carpalia are present and still in their original position, but they are so fragmentary that nothing can be said about them. Ileum. Both ilea are present. The left ileum is visible from the medial side. Its contour is complete, but most of the surface of the bone has broken off and portions of the upper part of the bone are only represented by impression. The right ilehm is badly preserved. It is visible from the ventral side and its position is more or less vertical to the plane of the remains. Therefore it has suffered greatly from crushing; The ileum consists of a large plate, which is slightly longitudinally convex as seen from the medial side, and which has a short anterior and a long pos- 84 Annals of the Transvaal Museum terior process, a long and slender preacetabular process and a short, thick postacetabular process. The upper portion of the ileum is very thin, the lower portion is thicker. The upper border is convex. The spina iliaca anterior terminates in a sharp point, while the spina posterior is broad and its end truncated. There is a low ridge on the medial surface of the hinder end of the ileum. This ridge starts behind the middle of the bone and is continued on to the spina posterior. It is situated very near to the lower border of the spina and is practically parallel with it. The hinder border between the spina posterior and the processus postacetabularis is straight. The length of this straight border is 21 mm. The open acetabulum cuts into the ileum to a depth of 17 mm. and a breadth of 41 mm. Above the acetabulum the medial surface of the ileum shows a deep fold, that is to say, here the lower portion of the ileum is bent inwards on the upper portion. This groove corresponds with the crista supraacetabularis on the lateral surface. 4The medial surface between this groove and the acetabular border is convex. The articulation surface may therefore be concave. Groove and acetabular border are nearest to each other at the root of the processus praeacetabularis. From this point border and Fig. 3. Aristosaurus erectus. Medial view of the left ileum. About nat. size. groove diverge backwards. From the position of the groove one would con- clude that the acetabular surface extends from the crista supraacetabularis inwards and very much downwards and therefore also forms a lateral articu- lation surface. The processus praeacetabularis is slightly narrower near its junction with the body of the bone than near its extremity. It reaches much further for- wards than the spina anterior. Between the two there is a deep and wide notch. The length of the processus praeacetabularis, measured from the base of this notch, is 36 mm. The distance between the end of the spina iliaca anterior and that of the spina iliaca posterior is 75 mm. The height of the ileum, measured from the lower end of the processus postacetabularis to the upper border, is 46 mm. Pubis. The pubes are beautifully preserved. The right pubis is complete, except for its lateral distal corner, of which, however, the impression is present. The left pubis is broken in two, the proximal half being in one slab and the distal half in the other. There is no indication that the pubes were fused together medially. This does not exclude the possibility that they were actually fused, because of the medial borders of both bones only a very small portion of that of the right pubis is preserved. Annals of the Transvaal Museum 85 The length of the right pubis is in mm. Its breadth at the downward fold is 27 mm. The narrowest part of the pubic plate, which is near its distal end, measures 17 mm., while the breadth of the distal end is 20 mm. The lateral border of the pubic plate is thick and its medial border is paper thin. The right pubis does not show the thickening of the distal end, but in the left one it is clearly visible. The whole lateral border of the pubis is concave. The medial border is straight. The upper surface of the pubic plate is concave, at least near its proximal end. The left pubis shows plainly that lower surface of the pubic plate is convex. The proximal medial corner of the pubic plate is bent downwards and at right angles to the larger portion of the bone. The angle is rounded. The pubic neck continues from the proximal lateral corner of the pubic plate to the articulatio ileo-pubica. The narrowest place of the pubic neck has a breadth of 11 mm. in the right pubis and a breadth of 10 mm. in the left one. It is situated at 22 mm. and 18 mm. from the articulatio ileo-pubica in the right and left pubis respec- tively. There is not much twist left in the neck of the right pubis, through the crushing of the bone. Its upper portion, however, still makes a considerable angle with the pubic plate. The medial side of the neck of the left pubis bends down more and more as it approaches the proximal end of the bone. The upper surface of the proximal end finally makes an angle of about 45 0 with the upper surface of the pubic plate. The upper border of the processus subacetabularis makes an angle of slightly less than 90° with the lateral border of the pubis. The breadth of the pubis across the processus subacetabularis is 25 mm. The height of the processus subacetabularis is much less at its lower than at its upper end. At its lower end the processus subace- tabularis is bent hooklike at an angle of about 90° in the direction of the distal end of the bone. The termination of the lower end is not abrupt; the upper border of the Fig. 4. Aristosaurus erectus. Upper processus subacetabularis rounds off gradually anterior view of right pubis, into the lower border of the hooklike process. About nat. size. Continuing distally, this process bends gradually more and more to the medial side, until it unites with the extreme end of the bent down proximal medial corner of the pubic plate. The breadth of this process at its middle is between 2 and 3 mm. Nothing is visible of the upper surfaces of the proximal end of the pubis. The neck of the pubis, the processus subacetabularis, the hooklike process and the proximal end of the pubic plate surround a large pubic foramen. It is situated nearest to the lower border of the proximal end. Its shape differs slightly in the two bones. In both it is elongated, and has a broad upper end. In the right pubis the lower end of the foramen is sharply pointed and the whole foramen pear-shaped; the upper end of this pear lies 86 Annals of the Transvaal Museum between the lower end of the hooklike process and the bent down proximal medial corner of the pubic plate; its broad base lies more proximally, in the direction of the middle of the upper border of the bone. In the left pubis the lower end of the foramen is not so sharply pointed. The greatest breadth of both foramina is 8 mm. In the right pubis this breadth is situated at about 4 mm. from its broad end, whereas in the left pubis this breadth is retained for a distance of 8 mm. from the broad end. The total length of the foramen is 1 8 and 21 mm. in the left and right pubis respectively. Ischium. Both ischia are preserved. Their condition, however, is very poor. Por- tions of the two bones are in both slabs. The contour of the right ischium is complete. The length of the bone is 102 mm. The bones are of course flattened, but the right ischium still seems to show that the and that the proximal plate as a whole is slightly bent outwards. The upper anterior edge of the plate is bent out farther than the upper posterior corner or the lower anterior edge of the plate. The proximal plate has a total breadth of 43 mm. Its posterior portion is thicker than its anterior portion. The medial surface of the proximal plate is concave an- teriorly; posteriorly the surface is not pre- served. The breadth of the process is 13 mm. At the anterior border the process is thin and at the posterior border it is thick. Distally the posterior border is straight; its proximal half is concave, except for a small portion near its upper end, which is convex. As the two bones were separated from each other before fossilisation and both of them are present in their full length, they cannot have been co-ossified at their distal end. The anterior border of the process is fairly straight. Just below the anterior border of the proximal plate it is concave. This point lies 37 mm. below the junction of the posterior with the anterior portion of the upper border. The anterior border of the proximal plate is very slightly convex. Proximally it bends gradually into the anterior portion of the upper border. There is no processus subacetabularis ischii. The upper border of the ischium consists of an anterior and posterior portion, which meet each other at an obtuse angle. The posterior portion makes an acute angle with the posterior border of the ischium. It is fairly straight, but at its posterior end it is slightly concave. This con- cave part corresponds with the articulation surface for the ileum. The anterior portion is straight in its upper part and rounds off into the anterior border of the ischium. The posterior portion of the upper border has a length of downward process is straight ischium, with impressions of its lateral side. About nat. size. Annals of the Transvaal Museum 87 26 mm. The impression of the lateral side of the bone shows some grooves and the bone must therefore have had fine ridges on this surface, indicated by the lines in fig. 5. A reconstruction of the pelvis shows a- remarkable feature. If the three bones are fitted together with the upper proximal corner of the pubis against the end of the processus praeacetabularis and the upper posterior corner of the ischium (its articulation surface) against the end of the processus post- acetabularis, then pubis and ischium can meet in such a way that the lower proximal corner of the pubis rests against the anterior portion of the upper border of the ischium, where the upper border rounds off into the anterior border of the bone. In this position both ischium and pubis would be greatly directed downwards and only slightly forwards and backwards if the line connecting the spinae is placed horizontally. However, the concave upper medial border of the pubis and the convex upper anterior border of the ischium suggest, in the light of the fact, that for example in Gryponyx africanus (4, p. 297) pubes and ischia form a complete symphysis right through, that in our case the connection between the two bones may have been greater. In order to find out whether this suggestion lay within the bounds of possibility, models were made in plasticine of the preserved shapes of the bones. As the thickness of the bones would have to be deduced by comparison with com- pletely known bones and as the thickness would not really affect the general argument, these plasticine models were made of uniform thickness. A recon- struction of the pelvis with the help of these models now showed, that if the ilea were placed so, that they diverged forwards and also slightly downwards and the pubes in a plane, it was possible to place the ischia in such a way that their upper anterior border nearly fitted into the concave upper posterior border of the pubes. There would be a complete fit if the lower anterior portion of the proximal plate of the ischium were bent slightly more inwards and the upper posterior corner of the pubis slightly more outwards. The whole fossil has suffered greatly from pressure and both ischia and pubes are lying flat in the slabs. It is therefore nearly certain that they have lost at least some of their original curvature. We may therefore conclude, that in this animal pubes and ischia had very probably a complete suture, from the distal end of the pubes to the distal end of the ischia. The distal ends, of both bones would then come still nearer to each other than was already assumed above. Femur. Both femora are present. The left one shows its hinder aspect, but only an impression of the front side of its distal end. The right femur is seen from the medial side. The bones are badly preserved, for both are more or less split in two and very much flattened. The length of the left femur is 137 mm. The breadth of the proximal end, measured from the tip of the caput femoris to the lateral end of the proximal surface, is 33 mm. Its thickness as shown in the right femur is 21 mm. The caput femoris projects far inwards. Its height at the medial end is 10 mm. At the extreme medial lower end of the caput there is a very small hooklike, downward process, which may perhaps be interpreted as the section of a rim along its lower border. The line of the proximal surface makes nearly a right angle with the lateral line. There is a slight thickening near the upper edge of the left femur, which may perhaps indicate the position of the trochanter minor. The centre of this thickening is situated at about 13 mm. from the medial end of the caput femoris. The trochanter major is not visible. Al- 88 Annals of the Transvaal Museum though very much damaged, the trochanter quartus is still recognisable. As preserved, the distance of its upper end from the proximal end of the bone is 41 mm. In the perfect specimen this distance may have been 2 or 3 mm. shorter. The trochanter is too much damaged to measure its length. It is situated much closer to the medial than to the lateral side. The lower end of the trochanter is further away from the medial side than the upper end. The fossa intercondyloidea is not visible. The anterior surface of the distal end is slightly concave in the middle. The border of the bone between the anterior surface and the distal surface is also slightly concave. The length (antero- posterior measurement) of what may be the condylus lateralis, measured from the anterior surface to the extreme tip of the condylus, is 33 mm. The section through the distal surface shown by the right femur does not show any obliquity to the axis of the bone ; only the posterior end of this line is slightly elevated. The breadth of the distal end is 28 mm. The shaft of the femur has a minimum breadth of 17 mm. in the middle. The minimum thickness of the bone is 16 mm., also in the middle. As seen from behind the femur is straight, but the lateral view of the right femur shows a pronounced sigmoidal shape. Tibia. Both tibiae are preserved, but the surface of the bones is badly damaged. They both show their medial side. The proximal end is very large and the shaft, as seen sideways, is straight. The length of the left tibia (see fig. 6) is 126 mm. ; that of the right one 129 mm. The length of the head is 36 mm. in the left and 38 mm. in the right tibia. The line of the proximal surface, which passes from behind upwards and forwards, curves gradually into the line of the anterior surface in the left tibia. In the right one, which is plainly damaged at this point, these two lines meet at an acute angle. Considering both bones, it seems that the tuberositas tibiae did not protrude much. Distally the proximal end becomes rapidly thinner and passes into the shaft. No thick- ening is visible of the distal end. The medial border of the distal surface is straight and at right angles to the shaft. The thickness of the distal end of both bones is 15 mm. The thickness of the shaft remains practically the same as that of the distal end till past its middle, where the shaft becomes gradually thicker towards the proximal end. The line of the anterior surface of the bone is very slightly concave and the line of the posterior surface is strongly concave in its proximal half. Fibula. Only the right fibula is visible. It is seen from the front, but its whole surface is very much damaged. Its length is 122 mm. The bone is slightly convex outwards. The breadth of the bone in the middle is 7 mm. and at its lower end 17 mm. The proximal end is partly covered by the tibia. The line of the distal surface is at right angles to the line of the lateral surface. Tarsalia. The right tarsus is complete and remains of the left one are also present. Of course all the bones are flattened and only show one side. The astragalus of the right foot has a length of 29 mm. and a breadth of 12 mm. It shows probably its upper surface, but this has only partly been preserved. Its hinder border is thin. The upper surface is divided into three concave portions by two feeble ridges, which start at the front border, about 8 mm. apart and diverge backwards. Annals of the Transvaal Museum 89 That which is still present of the calcaneum is triangular. The edge which is turned towards the astragalus has a length of 10 mm. The front edge has the same length and is slightly concave. The lateral angle seems to be more pointed than the two inner ones. The distal tarsalia probably also show their upper surface. The cuboid is the largest of the three. Its length is 15 mm. and the breadth of that of the left foot is 10 mm. It has a triangular shape and the line of what may be its hinder surface is convex and forms the longest side of the triangle. In the right foot it turns its sharpest angle outwards, but in the left inwards; how- ever, in the left foot the tarsalia seem to be displaced (see fig. 6). Cunei- forme III and II are both too deep in the matrix to say anything about them with certainty. Foot. Both feet are preserved. All the bones are flattened, although the right foot is slightly better in this respect than the left one. An impression with part of the bone of the left foot is present in both slabs. Metatarsal I is not visible in the right foot; the left one has a length of 38 mm. and distally a breadth 6f 14 mm. Its proximal end seems to be narrow. The lateral side of the distal end of the flattened bone is broad and rounded, while its medial side forms an angle at a short distance from the extreme distal end. This corresponds, therefore, with a large rounded lateral head and a small medial one as described by v. Huene for example in Plateosaurus Reinigeri (3, p. 28). As in that species, the axis of the articulation surface for the first phalanx would, therefore, not be at right angles to the axis of the metatarsal, but turned slightly inwards. This is actually the case, for the distal end of the first phalanx stands off inwards.* Metatarsal II has a length of 59 mm. Its proximal end is cut off straight. The lateral side of this end is hollow and forms a sharp ridge with the flat hinder surface. The medial side is not visible, but is probably also hollow to receive the proximal end of metatarsal I, which partly covers it. The length of the hinder border of the proximal surface is about 10 mm. The shaft is straight and has a width of 9 mm. at its narrowest part. The distal end is also cut off straight. It cannot be made out whether the axis of the distal articulation surface was at right angles to the axis of the bone or not. Pro- bably, however, its medial end was slighty higher than its lateral one, for the axis of the first phalanx deviates slightly inwards from the direction of the axis of metatarsal II. The distal end has a breadth of about 15 mm. Metatarsal III has a length of 66 mm. It is more slender than metatarsal II. Its proximal end has a breadth of 17 mm. and shows on its hinder surface a slight longitudinal ridge, possibly marking the extent to which it covered metatarsal IV. The shaft is straight and measures at its narrowest part 9 mm. Near the distal end the medial side of the shaft is concave and the lateral side convex. The breadth of the distal end is 14 mm. The axis of the distal articulation surface is not at right angles to the axis of the bone, as its medial end is slightly higher than its lateral one. Metatarsal IV has a length of 60 mm. It is a much more slender bone than metatarsal III. The breadth of its distal end is 12 mm. The distal arti- culation surface is not at right angles to the axis of the bone. Its lateral end is higher than its medial end. It rounds off gradually into the lateral side of the bone. The impression of its dorsal surface shows a broken off knob of matrix, which entered into the lateral collateral pit. Metatarsal V has a length of 30 mm. Its proximal end is broad and its 90 Annals of the Transvaal Museum distal end a blunt point. The proximal end has a breadth of more than 13 mm. Laterally on the front surface there is a ridge which runs towards the distal end. All phalanges are represented. Fig. 6. Aristosaurus erectus. Left foot seen from above. The first phalanx of the first digit shows its medial collateral pit; the other phalanges show the im- pressions of the deepened sides of their distal ends. Slightly reduced. I, 1 lies flat down on the slab, but shows its complete medial collateral pit and nothing of its lateral one. It may therefore be concluded that the distal end of the phalanx is turned slightly outwards. The claws are only slightly curved. The second toe deviates slightly inwards and the fourth slightly outwards from the third. Annals of the Transvaal Museum 9i Measurements of the phalanges in millimetres length prox. breadth dist. breadth 1st digit 1st phalanx 23 14 — 2nd 30 — — 2nd ,, 1st ,, 24 14 !4 2nd ,, 22 12 10 3rd 3rd ,, 1st ,, at least 26 24 15 13 2nd ,, 17 13 12 3rd 15 10 10 4th 4th ,, 1st at least 28 22 2nd 13 — IO 3rd 11 — 8 4th 10 — — 5th 5th ,, 1st ,, at least 23 7 — __ Discussion. The nearest relations of our form are amongst the Plateosauridae and Anchisauridae, and it is clear that it belongs to either the one or the other. Th ^^Plateosauridae are all medium sized to large Dinosaurs with fifteen dorsal vertebrae, whereas our form is small and has most probably fourteen dorsal vertebrae, agreeing in this respect with the Anchisauridae. There is further agreement with the Anchisauridae in the relation of the lower arm to the humerus; radius and ulna are longer than half the humerus. The length of the shaft of the humerus stands to the length of the whole humerus as 58 : 93 or 0-62. This relation brings our form in close proximity of Thecodontosaurus antiquus. Taking all these facts into consideration it seems clear that our form is an Anchisaurid. Comparison with Anchisaurus shows that the dorsal vertebrae are com- paratively longer, and that the pubes of the two forms differ greatly. Masso- spondylus is a much larger form. The distal ends of its ischia are coalesced, and each is more or less triangular in section. In our form the distal ends of the ischia are flattened through pressure, but it is difficult to accept that their section was originally triangular. Moreover, they are not coalesced. The re- lations of the ileum of Massospondylus carinatus (2, PI. XV) are different from those in our form, for it is longer than the latter with regard to its shortest height above the acetabulum. Relatively the dorsal vertebrae of our form are longer than those of Massospondylus carinatus. The relation of the lengths of the metatarsals in Massospondylus Harriesi (4, p. 303) is different from that in our form. In Massospondylus Harriesi metatarsal II is longer than meta- tarsal IV, whereas in our form metatarsal II is shorter than metatarsal IV. In comparing with A mmosaurus (3, p. 304) and Gyposaurus (1 and 4, p. 293) I need only refer to the great difference in the ilea. The only other genus of the family is Thecodontosaurus. Superficially there is great resemblance between our form and the known species of Thecodonto- saurus. A closer study, however, reveals remarkable differences. A comparison of the ileum of our form with that of Thecodontosaurus antiquus (3, Pis. LXXXII and LXXXIV) shows that in the latter the spina posterior is much more produced. The acetabulum cuts deeper into the ileum of our form, which resembles the Plateosauridae in this respect. The highest point of the acetabular concavity is situated much nearer towards the middle of the bone than in our form, and this is another point of resemblance with the Plateo- sauridae. The ileum of our form is manifestly different from that of Theco- dontosaurus cylindrodon (3, PI. LXXXIV), and also in the direction of the Plateosauridae. 92 Annals of the Transvaal Museum The pubis of our form differs considerably from that of Thecodontosaurus antiquus (3, PI. LXXXV), as far as the latter is known. In our form the lateral edge of the pubis is regularly concave, whereas in Thecodontosaurus antiquus its upper end is sigmoidal. There is also great difference in the shape of the pubic foramina. The shape of the proximal end of the ischium of Thecodontosaurus antiquus , as far as preserved (3, p. 209, fig. 228), is quite different from that of our form, a difference which is best understood from a comparison of the figures. Another difference becomes conspicuous when the length of the humerus is expressed in lengths of dorsal vertebrae. Taking one of the hinder vertebrae v. Huene came to the following results (3, p. 309) : In Thecodontosaurus anti- quus the humerus is about five times as long as the vertebra, and in The- codontosaurus skirtopodus about four and a half times. In our form the length of the eleventh dorsal vertebra is 29 mm. The length of the right humerus is 93 mm., which means that the humerus is only 3*2 times as long as the vertebra. Therefore, the humerus of our form is relatively much shorter than that of Thecodontosaurus antiquus and of Thecodontosaurus skirtopodus. There is great difference between the ischium of Thecodontosaurus minor (5, p. 469) and that of our form. The points of difference enumerated above show sufficiently that our form does not belong to any of the known genera of the Anchisauridae. It, therefore, represents a new genus, for which I propose the name Aristosaurus n.g. with the species Aristosaurus erectus n. sp. Aristosaurus erectus is much more highly specialised than Thecodontosaurus, Ammosaurus, Anchisaurus and even than M assospondylus . The build of the pelvis, and especially the position of the ischium, shows adaptation to a usually bipedal mode of locomotion. The same may be concluded from the far forward position of the acetabular concavity in the ileum. The position of the trochanter quartus seems to be very low down on the femur. Its upper end is 41 mm. from the proximal end of the bone. The length of the trochanter is at least 18 mm. Therefore the lower end of the trochanter is situated at more than 59 mm. from the proximal end of the bone, which means very near to the middle of the femur. However, conclusions may not be drawn from this fact, because exact measurements cannot be obtained. The humerus is much shorter in relation to the body than in the other Anchisauridae. The anterior extremity is also relatively much shorter in relation to the posterior one than in all other Anchisauridae excepting Anchisaurus solus. As in the Plateosauridae the tibia of Aristosaurus is much longer than the humerus. This is also the case in Anchisaurus solus. In the other Anchisauridae it is the reverse. All this tends to show that Aristo- saurus is an Anchisaurid, highly specialised in the direction of the Plateo- sauridae, and of the bipedal mode of locomotion. Literature. 1. Broom, R. On the South African Dinosaur ( Hortalotarsus ). Trans. S. Afr. Phil. Soc. vol. xvi. pp. 201-206, PI. hi. Cape Town, 1905. 2. Huene, F. von. Ueber die Dinosaurier der aussereuropaeischen Trias. Geol. u. Pal. Abh., herausg. v. E. Koken, N.F. Bd. vm. H. 2. Jena, 1906. 3. Huene, F. von. Die Dinosaurier der europaeischen Triasformation. Geol. u. Pal. Abh., herausg. v. E. Koken, Suppl.-Bd. 1. Jena, 1907-1908. 4. Broom, R. On the Dinosaurs of the Stormberg, South Africa. Ann. S. Afr. Mus. vol. vii. Pt. iv. No. 13, pp. 291-308, Pis. xiv-xvii. London, July 5th, 1911. 5. Haughton, S. H. A new Dinosaur from the Stormberg Beds of South Africa. Ann. and Mag. Nat. Hist. Ser. ix. vol. 2, No. 12, pp. 468-469. London, Dec. 1918. Annals Transvaal Mus Vol. VII, Pt 2 Plate IX Aristosaurus erectus. Slab on which the animal lies with its ventral side downwards. This therefore gives a dorsal view of the animal. 0*193 x nat- size. | Annals Transvaal Mus. Vol. VII, Pt. 2 Plate X The covering slab of the one in PI. IX. This gives a ventral view of the animal and the right side of the plate corresponds therefore with the left side of the animal. 0*214 x nat. size. Aristosaurus erectus. Annals of the Transvaal Museum 93 CONTRIBUTIONS TO THE KNOWLEDGE OF THE REPTILES OF THE KARROO FORMATION By Dr E. C. N. van Hoepen, M.I. 6. FURTHER DINOSAURIAN MATERIAL IN THE TRANSVAAL MUSEUM. With 13 plates and 27 text-figures. Eucnemesaurus fortis n.g. et sp. The remains to be described under this name were collected by myself from a red mudstone, probably belonging to the Red Beds, on the farm Zonderhout, near the railway station of Slabberts. High above the locality is a krans of, probably, Cave Sandstone. The remains consist of the proximal half of a femur, a complete tibia, a proximal portion of a pubis, portions of dorsal and caudal vertebrae and some fragments. Dorsal Vertebrae. There are portions of three dorsal vertebrae. One centrum belonging to the region between the tenth and the fourteenth, a portion of a neural arch, which belongs to one of the vertebrae from the seventh backwards, and a neural arch which probably belongs to the sixth vertebra. The neural arch of the sixth vertebra is very much weathered. The anterior supporting ridge is narrow and in its turn supported by a smaller ridge, which starts near its upper end, runs downwards and nearly parallel to the posterior supporting ridge and disappears before reaching the centro-neural suture. There are deep cavities on both sides of this ridge. The sutural surface of the centro-neural suture shows a transverse groove in the middle. The breadth of the neural canal is 1*5 cm. posteriorly and 2 cm. anteriorly. Its height above the sutural surface is 2-5 cm. In the other neural arch there is probably only one supporting ridge for the processus transversus. It is too fragmentary to merit detailed description. The centrum (PI. XII, figs. 3 and 4) is large. Its length is about 10-5 cm. It has suffered from pressure, for one side is longer than the other and its lower surface shows signs of compression. The height of its anterior and posterior articulation surface is 1 1 - 5 cm. The breadth of the anterior surface is 12 cm. and of the posterior surface it may have been 13 cm., but this cannot be measured, because the right border of the posterior end is broken off. The posterior articulation surface is more concave than the anterior one. There is a deep longitudinal groove on both sides of the centrum and above its middle. There is a large knob on the upper anterior corner of the side of the centrum; this is a portion of the anterior supporting ridge of the processus transversus. The thickness of this knob suggests that this vertebra may be the thirteenth or the fourteenth. The articulation surfaces of the centro-neural suture are broad. A ridge runs transversely across their middle. The breadth of these surfaces in the middle is 3 cm. The neural canal is very deep in the middle of the vertebra. The bottom of the deepest part is divided in two by a ridge. The breadth of the neural canal in the middle is 1*3 cm. Its depth below the centro-neural sutural surface is 3 cm. 7 94 Annals of the Transvaal Museum Caudal Vertebrae. Remains of four caudal vertebrae are present (PI. XIII, fig. i). One of these is apparently the greater part of the fourth caudal. Its distal articulation surface is broken off. The concave anterior articulation surface has a height ! of 9 cm. and a breadth of 8-5 cm. Its lower border is much thicker and broader than the sides; this is for articulation of the haemapophysis. The centrum is thickest immediately below the transverse processes, where it measures 5 cm. The sides converge downwards, and as far as can be made out, the lower surface is rounded. If the anterior articulation surface is placed vertically, the upper surfaces of the broken off transverse processes slope backwards and downwards. The praezygapophyses are broken off. The anterior edge of the processus spinosus starts somewhat in front of the middle of the bone and slopes upwards and backwards. This front edge is concave. The section of the base of the process is wedge-shaped. The base of the transverse process is very broad. It starts at about 3 cm. from the anterior edge of the centrum. Its thickness on the left side is i*8 cm. Two others are from the middle caudal region and perhaps from the twelfth to the sixteenth. The posterior one of the two has lost its anterior half and its dorsal spine. Its concave posterior articulation surface has a height of 6 cm. and a breadth of 5 -5 cm. The hinder border of the left transverse process lies 2 cm. in front of the hinder border of the centrum. The anterior one of the two has lost a piece of its anterior articulation surface. The length of its centrum is 7-8 cm. above and 6*2 cm. below. The anterior articulation surface is more concave than the posterior one, but this may be due to the developing, because there was scarcely any difference between the bone and its incrusta- tion. The transverse process is attached about 0-5 cm. in front of the posterior border of the Centrum. The height of the anterior articulation surface is 6 cm. The height of the posterior surface is slightly less and its breadth 5-3 cm. The fourth vertebra belongs to the end of the middle caudal region. It is too much weathered to give any detail. The length of the centrum was at least 6 cm. and its height at least 3-5 cm. There is a transverse process which has a proximal breadth of 2 cm. The processus spinosus stands on the distal half of the bone. Pubis. Two pieces of the left pubis are preserved, the distal and the proximal end. The proximal end (PL XI, figs. 3 and 4, and text-fig. 1) consists of the neck with the processus subacetabularis. The distal end of the piece still belongs to the pubic plate. The inner side of this end is bent down. The section at the distal fracture is a flat oval, the inner end of which is elongated and ulti- mately broken off. The neck gives the impression of being peculiarly short. Its section differs greatly from that of the pubic neck of Dromicosaurus gracilis described hereafter. Through lack of material and literature it could not be made out whether this difference is of family value. There is no demarcation between an anterior and a medial surface of the neck. A section shows a regularly convex line from the medial border of the neck over the medial and anterior surfaces to its lateral border. The neck has a distinct lateral surface, which broadens rapidly into the lateral surface of the proximal end, and which narrows down rapidly into the lateral border of the plate. This lateral surface of the neck rounds off broadly into the hinder surface, which is concave in all directions. The greatest breadth of the neck, from the inner to the lateral border, is 6 cm., while its greatest thickness, measured at the upper fracture (see PI. XI), is 3 cm. The inner portion of the neck is thin and bent backwards. Annals of the Transvaal Museum 95 The lateral border of the neck is slightly concave forwards. This border broadens out proximally into the broadly rounded upper anterior corner of the bone. The medial surface of the upper end is convex antero-posteriorly, but concave from above downwards. Its lateral surface is concave. The arti- culatio ileo-pubica is broad and slightly twisted. The upper anterior portion Text-fig. i. Outline of transverse section through left pubic neck of Eucneme- saurus fortis in the region of the proximal fracture (PL XI) . Seen from the distal end. The anterior border is above and the medial border on the left-hand side. Nat. size. The slight concavity in the lower border deepens rapidly towards the distal end. Text-fig. 2. Outline of transverse section through left pubic neck of Dromicosaurus gra- cilis near the upper end of the foramen obturatorium. Seen from the distal end. The anterior border is above and the medial border on the left-hand side. Nat. size. of the lateral border of the proximal surface is higher than that of the medial border, while the lower posterior portion of the lateral border, just before reaching the groove, is lower than the corresponding medial border. The processus subacetabularis is broken off through, the groove on the proximal surface. This groove has a depth of 1*5 cm. and does not reach the medial border. The medial border of the proximal surface is evenly convex, while the lateral border is slightly concave. One gets the impression from the proximal end that its posterior portion is slightly twisted to the lateral side. The fractured end of the processus subacetabularis shows a convex lateral and a slightly concave medial border. Taken generally, the section at this end is triangular. The distal end (text-figs. 3 and 4) is very much weathered. A portion of the rounded lateral side is still visible. Its breadth as preserved is proximally 9 cm. and distally 7-5 cm. Its thickness is 3-5 cm. The greatest breadth of the distal surface as preserved is 4-5 cm. The medial -hinder-lateral border of the distal surface forms practically half a circle. Its front border is straight. The distal surface makes an acute angle with the anterior surface. I wish to lay stress on the fact that the anterior surface is the only one which is not weathered. Femur. Only the proximal half of the left femur is preserved (PI. XII, fig. 1 and text -fig. 5). The piece is broken off through the trochanter quartus. The 7—2 96 Annals of the Transvaal Museum Text-fig. 3. Eucnemesaurus fortis.. Outline of distal surface of left pubis. The anterior bor- der is below, the lateral border to the left, x A. length of the bone as preserved is 26-5 cm. The breadth of the proximal end, measured from the medial end of the caput femoris to the lateral side, is 15-5 cm. Its maximum thickness, measured medially to the trochanter minor, is 9 cm. The caput femoris is directed inwards; it has a breadth of 8-5 cm. and a height of 8-5 cm. The proximal surface is convex, rough and pitted, and forms an angle with the lateral side. The trochanter minor lies on the hinder surface and very near to the border of the proximal surface. It is a thick broad boss, the middle of which is situated at 8-5 cm. from the medial end of the caput femoris and at 6*5 cm. from the lateral side of the bone; in this therefore it differs from the trochanter minor of most Plateosauridae y which is there situated much nearer to the medial than to the lateral side. The height of the trochanter minor is 1 cm., its length is about 6 cm. and its breadth about 5 cm- The upper end of the trochanter major is situated at a distance of 12 cm. from the proximal end of the bone. The ridge of the trochanter is worn away, but it is still plainly visible that its lateral side formed a deep, broad groove with the general surface of the bone. The height of the trochanter near its proximal end as preserved is 2*5 cm.; it may originally have been 1 cm. higher. The length of the trochanter is about 10 cm. The femur is broader and thinner between its proximal end and the trochanter major than further downwards. A broad thick ridge runs from the trochanter major downwards apparently in the direction of the condylus medialis. This ridge continues upwards from the trochanter, but diminishes gradually in height and loses itself at about midway between the trochanter and the proximal end of the bone. The ridge is concave on its medial side opposite the trochanter major. The trochanter does not lie on the top of the ridge, but is situated on its lateral side. The medial side of the anterior surface of the bone is concave as far as preserved. The lateral side of the anterior surface, below the trochanter major, is convex. The medial side of the proximal end of the bone is concave, while its lateral side is slightly convex. Therefore, the upper end of the bone is turned slightly inwards. The hinder surface of the proximal end was originally flat; now it is irregularly concave through pressure. Opposite the trochanter quartus it rounds off, over the lateral side, into the anterior surface. Higher up, the lateral side is flatter and its borders with the anterior and posterior surfaces, are more conspicuous. The trochanter quartus begins at a distance of 17-5 cm. from the proximal end of the bone. Only a small portion of the trochanter is present; the lower half of the trochanter and the whole of its upper edge are lost. The portion preserved has a maximum height of 4 cm. Its breadth at the base, just above the fracture, is about 5-5 cm. The medial side of the trochanter is concave and the lateral side convex. Nothing is visible of a second upper ridge, but this may be due to crushing. The trochanter is situated nearer to the medial than to the lateral side of the bone. The breadth of the diaphysis over the Text-fig. 4. Eucnemesaurus fortis _• Outline of longitudinal sec- tion of distal end of left pubis. Distal border to the right, anterior border below, x i. Annals of the Transvaal Museum 97 trochanter is 9 cm. Its thickness, measured from the top of the upper end of the trochanter quartus as preserved to the top of the ridge on the anterior surface, is 12 cm. The section of the bone, which is revealed at the fractured end, is re- markable. The fractured surface has been polished and its photcfgraph is given in fig. 2 of PI. XII. It shows a thick wall of substantia compacta which is clearly lamellar. Sharply defined from this is an inner mass of substantia spongiosa. Only a small portion of the whole section is not occupied by bony matter. Here we have, therefore, a Theropod femur of practically solid structure. Zittel in his “Handbuch” (1) states on p. 718 that the limb-bones of all Thero- poda are hollow and were perhaps filled with air. In Eastman’s Text- book it is stated in the diagnosis of the Sub-Order Theropoda (3, p. 227) that the limb-bones are hollow, v. Huene, in his great work “Die Dinosaurier der europaischen Trias- formation,” compares the Theropoda with the Sauropoda on p. 346, and says: “The limb-bones of the Sauro- poda are not tube-like, but nearly completely massive, whereas those of the Theropoda are hollow.” In the second edition of Zittel’s handbook in 191 1 , Broili states in the diagnosis of the Sub-Order Theropoda that the limb-bones are mostly tube-like. Probably, therefore, Broili knew of Theropod limb-bones which were not hollow, and our femur now adds to the evidence. Tibia. The left tibia (PI. XI, figs. 1 and 2, and text-figs. 6 and 7) is complete. It has a very large proximal articulation surface and a straight shaft. The bone was found in five pieces, which all fitted neatly together. The total length of the bone is 46 cm* The head of the tibia has a length of 19-5 cm., measured from the hinder end of the medial condylus to the anterior edge of the tuberositas tibiae. Its greatest breadth, transverse to this grea^st length, is 13-5 cm. The length of the lateral side of the head is about 15 cm. An- teriorly the head of the tibia has a laterally directed process, the tuberositas tibiae. The lateral border of the proximal articulation surface is slightly con- vex at the tuberositas, more convex around the lateral condylus and sharply concave in between. The medial border runs from the tuberositas tibiae in- wards and curves gradually backwards; the posterior part of this border is not so convex as the anterior portion. The posterior border is convex behind each condylus and notched in the middle ; this notch is still filled with matrix. Text-fig. 5. Eucnemesaurus fortis. Outline of the posterior view of the proximal end of the left femur, showing the position of the trochanter minor and the upper end of the trochanter quartus . x 4. g8 Annals of the Transvaal Museum The proximal articulation surface is roughly convex, with small concavities in its anterior portion. The lateral posterior side of the surface is lower than the medial posterior corner and this again is slightly lower than the anterior end. The hinder border is overhanging. The shape of the proximal articulation surface is given in text-fig. 6. The shaft narrows down quickly immediately below the head. The distal end is transversely broad. The medial one-third of the distal articulation surface is more or less flat and slopes very slightly from its anterior edge backwards and downwards. The anterior part of this surface is convex. The lateral part of the distal articulation surface consists of an anterior and a posterior portion. The anterior portion, which is the distal surface of the pro- cessus tibiae anterior distalis, occupies a much higher position than the posterior Text-fig. 6. Eucnemesaurus fortis. Left portion, which is the distal surface of the t‘bia- Outline of proximal articula- r . , ,. . .. /-p.1 tion surface seen from above, processus tibiae posterior distalis. The Laterai side to the left, x J. distal surface of the processus posterior is continuous with the medial portion of the articulation surface. It slopes upwards from the lateral to the medial side and from the anterior towards the posterior border. As the processus posterior is rather weathered, especially laterally, the last fact might be due to this cause. If it is original, it would be a marked difference with the tibia of Dromicosaurus gracilis, where the sur- face slopes upwards from the posterior towafds the anterior border. The medial end of the distal surface of the processus anterior slopes down gradually towards the medial portion of the articulation surface, with which it j oins up by means of a broadly convex ridge. The distal face of the processus anterior slopes from its medial end upwards and outwards and from its anterior border backwards and slightly upwards. Medially the processus posterior is slightly broader than the processus anterior. It is bluntly pointed towards the lateral side. Laterally it is bent very slightly forwards. Both ends of the distal surface of the pro- cessus anterior are broad, but the inner end is slightly broader. The lateral end of the processus does not project further outwards than the processus posterior. The highest point of the lower surface of the processus anterior is situated 46 mm. above the lowest part of the processus posterior. The sulcus malleoli tibiae is very deep. This sulcus begins rather far above the processi and passes down between them towards the hinder border of the processus anterior. The greatest breadth of the distal end, parallel to the front border, is 14*5 cm. The front border is laterally slightly concave and medially convex. The medial border makes an acute angle with the anterior border and an obtuse angle with the posterior border. The anterior medial corner is globose; Text-fig. 7. Eucnemesaurus fortis. Left tibia. Outline of distal articulation surface seen from below. Lateral side to the left, x J. Annals of the Transvaal Museum 99 its medial and anterior sides are slightly overhanging. The medial border rounds off into the posterior one. The greatest breadth of the medial side is ii cm. The posterior border converges slightly with the anterior border to- wards the lateral side; its length is io cm. The distance from the anterior medial corner of the distal end to the lateral end of the processus posterior is nearly 16 cm. A broad high ridge runs down from the tuberositas tibiae in the direction of the anterior medial corner of the distal end, but disappears in the lower half of the bone. The lateral anterior edge of the distal end is broadly rounded below. Higher up the edge becomes a fairly sharp ridge, which disappears above the middle of the bone. There is a broad low boss on the lateral side of the bone, below the notch in the lateral border of the proximal surface. Its highest point lies about 8-5 cm. below the notch. The surface of the bone being generally crushed, no further particulars can be given. The thickness of the bone in the middle is 7 cm. and its breadth at this place 5-5 cm. Both dimensions were originally somewhat larger. The thickness of the distal end is slightly more than 9*5 cm. At the middle fracture the bone shows a thick wall of substantia compacta but no substantia spongiosa. The central cavity is fairly large. There are three fragments of the same specimen which could not be identi- fied with certainty. One of these is probably a piece of the shaft of a femur. Another piece might be identifiable if better material were present for com- parison. The third piece is an end of a large bone. The surface at the end has a length of 17-5 cm. and a breadth of 11 cm. At one end the sides of this surface meet at an acute angle; at the other end its border is broadly rounded. The shaft becomes rapidly narrower and at a distance of 10 cm. from the end its breadth is 8-5 cm. Its thickness at this point may be about 5 cm. The only bones it could belong to are the ischia and the fibulae. For the distal end of the ischium it is far too large, but for the distal end of the fibula it also seems too large. Its shape is exactly what one would expect of the distal end of the fibula. * Two other bones were found in the same locality, which belong to a much larger animal. One is a fragment of a tibia and the other of a coracoid. The tibial fragment was found loose on the same spot as the remains described above. It has probably broken away from higher strata. The coracoid was found loose on a very much higher level and was probably originally in the same conglomeratic bank as the tibial fragment. The fragment of the tibia is the lateral portion of the head of the right bone. It has a length of 22 cm. corresponding with about 13 cm. in the tibia of Eucnemesauvus . In comparison with the head of the Eucnemesauvus tibia the head under discussion may have had a total length of 33 cm. The coracoid consists only of the foramen supracoracoideum with some bone substance around it. It is apparently a right coracoid. The foramen passes upwards and inwards. Its outer opening is oval shaped, having a width of 3-5 cm. and a height of 2-8 cm. The inner surface below and behind the foramen is concave, the outer surface convex. The thickness of the bone immediately behind the foramen is 3 cm. It becomes thinner towards the scapular border. Immediately in front of the foramen the bone has a thickness of 6-5 cm. Both these bones indicate Theropodous Dinosaurs of tremendous size. Discussion. The described remains of the smaller animal show Plateosaurid characters, and especially is the tibia so typically Plateosaurid that I do not hesitate in 100 Annals of the Transvaal Museum placing the described form in that family. It will, however, be necessary to compare it with the different forms grouped together under this name and with those without fixed position; perhaps also with some Anchisaurid. The South African Plateosaurid Euskelesaurus is known by some frag- j mentary remains. Only a few of the bones can be used in comparison and among these the femur is the most important. Now the femur of Euskele- saurus Browni is incomplete, both ends being broken off and only the diaphysis with the trochanter major and trochanter quartus being present. Comparison is therefore rather difficult. In Euskelesaurus Browni the trochanter major lies 15 cm. below the present proximal end of the femur. From the figure given by v. Huene (4, p. 30) it is manifest that the upper end of the trochanter quartus lies from 26 cm. to 28 cm. below the proximal end of the bone. In our form the distances of trochanter major and trochanter quartus from the proximal end are 12 cm. and 17-5 cm. respectively. From these measurements can easily be deduced that in Euskelesaurus Browni the trochanter quartus lies relatively lower than in our form. Of course the relation will change if the missing portion of the head is added. Now one cannot say exactly how much of the head is missing, but let us for one moment suppose that the relations were the same in both bones. To get the same relations as in our form, 9 cm. would have to be added to the proximal end of the femur of Euskelesaurus Browni. The distance of both trochanters from the proximal end would then be 24 cm. and 35 cm. respectively; these distances would therefore be twice the same distances in our form. v. Huene estimates that not much more than 5 cm. is missing from both ends together, and 9 cm. is therefore far above his estimate. The distal end of the Euskelesaurus femur is missing and if some 10 cm. are added for the missing part the whole bone would then have a length of 9 + 65 + 10 = 84 cm. According to the above relations our femur would have to be half the size of the Euskelesaurus bone and therefore have a length of at the most 42 cm. That is at least 4 cm. shorter than the tibia ! On the other hand the lower end of the trochanter quartus in our form was situated at least 28 cm. from the proximal surface. This distance can .certainly have been greater. In Euskelesaurus Browni the lower end of the trochanter quartus lies about 40 cm. below the present proximal end, and if 9 cm. are added to this end, the distance of the trochanter quartus from it will be 49 cm. To get comparable results the same length as above must be maintained, viz. 84 cm. The lower end of the femur will therefore measure 35 cm. If the same relations existed in the two bones, the length of the portion of our femur below the lower end of the trochanter quartus would be 28 x 35 : 49 or 20 cm. The total length of our femur would then be at least 48 cm. Only one thing can follow from these different results for the length of our femur and that is, that our premise of the same relations in the two bones is wrong. The conclusion that the tro- chanter quartus of Euskelesaurus Browni lies relatively lower than the one of our form is right. The sections through and near the pubic neck of Euskelesaurus Browni (4, p. 29) are quite different from those of our form. There are ridges on the lower surface of the neck, which are absent in our form. The shape of the proximal surface of the pubis is also different. The head of the tibia of Euskele- saurus Browni has a greater lateral concavity. There seems also to be great difference in the distal end, but the figures given by v. Huene are not quite clear. According to the text (4, p. 32) the posterior border of the distal end of the tibia in Euskelesaurus Browni is much longer than the anterior border. In our form this is just the reverse, v. Huene mentions that the type of Annals of the Transvaal Museum ioi tibia of Euskelesaurus Browni also occurs in Gresslyosaurus robustus of Beben- hausen near Tubingen. The distal end of the tibia of Gresslyosaurus robustus is figured and described on pp. 134 and 135 of v. Huene’s great work on European Dinosaurs (5). In the explanation of fig. 133 it is stated that the lower border of the figure corresponds with the anterior border of the distal end ; this lower border, however, is much longer than the upper border, which would then correspond with the posterior border of the distal end. According to this figure, therefore, the anterior border of the distal end would be longer than the posterior border, as in all other Triassic Theropoda. Again, the ex- planation of fig. 134 states that the figure gives a postero-lateral view of the same distal end. From this figure it is apparent that the lateral end of the processus anterior is higher than that of the processus posterior and that both these processes correspond with the anterior and posterior border respectively of fig. 133. Now, if one compares the plate figure of this distal end on PI. LIV, which is given in front view, with the text-figures, it will be found that the processus anterior is identical with the processus anterior of the text-figures, further that the processus posterior projects further downwards than the pro- cessus anterior and that the whole posterior breadth of the bone is hidden behind the anterior breadth. The three figures are therefore identical among themselves and they show the same build of the distal end of the tibia as in all other Triassic Theropoda. The description in the text, however, is quite different. It is stated, that the posterior medial angle of the distal end is acute and projects further to the medial side than the anterior angle; in the figure, however, the posterior medial angle is obtuse, while the anterior one is acute and projects further to the medial side. It is also stated in the text that the hinder convex border has a length of 15-5 cm. However, the con- vexity and the dimension correspond with the anterior border of the figures. As the figures correspond with the general build of the distal end of the tibia in Theropodous Dinosaurs, I can only conclude that some mistake must have entered into the description. A comparison with Gryponyx africanus as described by Broom (7) is very difficult. It is of the greatest importance to compare the proximal end of the pubis of our form with that of Gryponyx, especially with regard to the shape of the neck, but from the description it is quite impossible to do so, for Dr Broom only states that “The upper end of the pubis is large and fairly flat/' and “It has a large nearly round pubic foramen.” Mr S. H. Haughton of Cape Town, however, has very kindly furnished me with the necessary in- formation. It appears that the pubic neck of Gryponyx africanus is very broad and flat, being nearly three times as broad as thick. Another very desirable comparison is that of the tibiae. The tibia of our form is typically Plateosaurid. This is especially shown by the shape of the head. Dr Broom describes the tibia of Gryponyx africanus as follows: “The tibia has a larger head than appears to be the case in Plateosaurus, but otherwise presents no special interesting features. The total length of the bone is 447 mm. and the proximal end measures 163 mm. by 71 mm.” That is all. Nothing about the shape of the proximal nor of the distal end. Only a few measurements to work with. It appears that our tibia is 1-3 cm. longer. Its head however is 3 cm. longer and 6 cm. broader. Therefore the tibia of our form has a very much larger head than that of Gryponyx africanus. A figure of the tibia of Gryponyx africanus has been given (7, PI. XIV, fig. 4) and in the explanation this is called an “outer view of left tibia.” In an outer view the sulcus malleoli should be visible. It is, however, not shown in the figure. It would be re- markable if this tibia did not possess such a sulcus. There is, however, no 102 Annals of the Transvaal Museum reason whatever to accept such a condition, for the figure is not an outer view of the left tibia, but an inner view of the right one. The following points will demonstrate this conclusion: A lateral view would not only show the sulcus malleoli but also the proximal articulation surface, because the latero- posterior corner is the lowest part of the head of the tibia in Plateosauridae and Anchisauridae. Neither of these two are visible in the figure. The border between the side figured and the distal surface is horizontal, as is the case with the medial border of the distal end of the tibia in all Triassic Theropods. In the figure the anterior border of the distal end passes from the anterior end of this horizontal border upwards and forwards (the anterior border of the proximal end is on the left-hand side) . In an outer view of the left tibia, in which the anterior surface of the bone is visible, the anterior border of the distal surface would pass from the left end of the horizontal medial border upwards and backwards. An inner view of the left tibia and an outer view of the right one would have their anterior surface on the right-hand side. Dr Broom does not give the distances of the trochanters from the proximal end of the femur. Taken from the figure, the upper end of the trochanter major lies at a distance of io cm., and the upper end of the trochanter quartus at a distance of 18 cm. from the proximal end of the bone. In our femur, therefore, the trochanter major is situated relatively lower than in Gryponyx africanus. A comparison with Plateosaurus stormbergensis Broom (9, p. 162) is practi- cally impossible from the “ descrip tion.” Dr Broom mentions three bones, the metacarpale, the femur and the pubis. Of the femur the length, the breadth of the distal end and the distance from the proximal end to the top of the median trochanter are given as principal measurements. Essentially the description of this bone consists of the statement that it agrees very closely with the femora of existing species of Plateosaurus. The existence of this close agreement is greatly appreciated, so much so, that one develops an irresistible wish to know exactly how this femur differs from those described before. However, to arrive at this knowledge, it is necessary to have among others another principal measurement, the distance of the trochanter major from the proximal end, which Dr Broom does not give. I conclude, from the figure of the femur of Plateosaurus stormbergensis (9, p. 164), that the proximal end of the femur in our form is more massive than in the type from the Stormberg. The head of the tibia of our form is relatively much broader than that of M assospondylus Harriesi. A comparison of the tibia of our form with that of the European Plateo- sauridae shows that the tibial head of our form is relatively much bigger than that of Plateosaurus Reinigeri and that the relations of the head of the tibia in Plateosaurus erlenbergiensis, Plateosaurus poligniensis , Gresslyosaurus robus- tus and in Pachysaurus ajax are quite different from those in our form. It follows from the above considerations that the present form is new and belongs to a new genus. I therefore propose to call it Eucnemesaurus fortis n.g. et sp. (ev = good, nv^pr] = tibia, o-avpos = lizard; fortis — strong). Gryponyx transvaalensis Broom. Through the kindness of Dr Broom the Transvaal Museum now possesses the type specimens of Gryponyx transvaalensis. These consist of a claw- phalanx and the distal end of a metatarsale. The claw has been described by Dr Broom as belonging to the first digit of the right manus (8, p. 82, fig. 3). The clawphalanx is high and strongly curved. Its right side is somewhat Annals of the Transvaal Museum 103 weathered, but it is still plainly visible that the groove for the claw on the right side lies slightly deeper than that on the left. There is a broad ridge on each side below this groove. The highest part of this ridge on the right side lies so far below its middle line, that it forms a fairly sharp lower edge to the right side of the bone. The highest part of the ridge on the left side .of the bone follows the middle line of the ridge. The ridge on the right side passes further backwards than that on the left. There is a broad, thick, boss in front of the middle of the articulation surface on the left side. That on the right side is broken away. This boss is separated from the ridge in front of it by a broad groove, which passes downwards and then backwards above the tuberositas for the flexor tendon ; upwards the groove is continuous with the groove for the claw. The greater part of the tuberositas for the flexor tendon is broken away. The articulation surface has completely disappeared by weathering. If compared with the clawphalanx of the first finger of Massospondylus Browni described hereafter, it will be noticed that what is called here right and left corresponds with medial and lateral respectively there. This therefore means, that the claw is not of the right manus as stated in the original descrip- tion, but of the left. I do not think that it belongs to the same species as Massospondylus Browni, because the proximal end of the lateral side ridge lies relatively much higher, with regard to that of the medial side, than the proximal end of the lateral side ridge in Massospondylus Browni. Whethet it belongs to the genus Gryponyx I am unable to say at present. Dr Broom mentions “the vascular groove” and one is led to believe that only one such groove is present. This, however, is not so, for although the medial side of the bone is weathered, one can still distinctly trace the groove on that side. In the description of Gryponyx africanus (7, p. 296) Dr Broom states of the clawphalanx of the first finger: “On the radial side is a marked vascular groove.” As the statement stands, one concludes that such a groove is not present on the “ulnar side.” However, I very much doubt the truth of such a conclusion and until it is contradicted I will accept Gryponyx africanus to have a clawphalanx of the first finger with two grooves for the claw, as in all other Triassic Theropoda. The distal end of the metatarsale has been described as that of the second bone of the right foot. The breadth of the distal end is 3-8 cm. and its thick- ness 2-4 cm. The medial end of the axis of the articulation surface is slightly higher than its lateral end. The middle of the posterior surface immediately above the articulation surface is concave. Laterally it is bordered by a ridge, which runs upwards from the lateral lower corner of the surface and practi- cally parallel to the lateral side. The medial posterior process of the distal end is short and directed about just as much outwards as inwards. The anterior surface shows a slight thickening above the middle of the articulation surface, bounded above by a very shallow, half-moon shaped concavity. On the lateral border of the anterior surface this concavity is bounded by a broad low boss, which forms at the same time the upper anterior knob-like process of the border of the lateral collateral pit. This pit has an upper, a lower and an anterior outlet. The section at the fractured end is broad laterally and narrow medially. Some of the characters of the bone, as for example the ridge on the posterior surface and the section of the shaft, point to a second metatar- sale, but others, as for example those of the anterior surface, point to a third metatarsale. At present I am unable to decide the point. Dromicosaurus gracilis n.g. et sp. The following remains were collected by myself from the Red Beds of 104 Annals of the Transvaal Museum Nauwpoort Nek, Bethlehem District. The exact locality is about ioo yards above the road from Bethlehem to Clarens as it passes through Nauwpoort Nek and on its right-hand side coming from Bethlehem. The remains consist of fragments of humerus and radius, a fairly complete neck-vertebra, some caudal vertebrae, the pubes, the ischia, a femur, a tibia, a fibula and some foot-bones. Neck Vertebrae. One neck- vertebra is preserved. It belongs to the anterior region of the neck and may possibly be the third vertebra (PI. XVI, fig. 3). A portion of the neural arch with the processus spinosus is broken off. Its left side is still partially covered by matrix. It is cracked in different places and a broad crack affecting the length is now filled with matrix. The length of the centrum as preserved is 7-6 cm. Height of its anterior articulation surface 4-7 cm. Height of its posterior articulation surface 5-1 cm. Breadth of the anterior and posterior articulation surfaces 4-6 cm. and 4-7 cm. respectively. The lower surface of the centrum is deeply concave, the deepest part of this con- cavity lying about 3 cm. behind the anterior surface; the height of the centrum is here about 3 cm. and its breadth about 2 cm. Both the articulation surfaces of the centrum are deeply concave. The left praezygapophysis projects far forwards and its articulation surface is perfectly horizontal. Caudal Vertebrae. There are remains of some eight caudal vertebrae (PI. XVI, fig. 1). They probably all belong to the anterior caudal region and are very badly preserved. The two biggest centra (PI. XVI, fig. 1 a) have a length of 5 cm. and a height of 6 cm. Their upper breadth is about 4*5 cm. The lower end of their distal articulation surface is strongly recurved for the articulation of the haemapo- physis. The lower surface of the centrum is narrow and fairly flat; there may have been a very slight groove, but it is not visible now. The next following in size (PI. XVI, fig. ib) is only represented by the upper half of its centrum with its dorsal spine. One praezygapophysis is preserved. It is long and its articulation surface slopes inwards and downwards. The postzygapophyses are short and situated high up behind the dorsal spine. Their articulation surfaces have the same slope as of the praezygapophysis. The end of the dorsal spine is broken off; as preserved, its length is 5-5 cm. It is narrow and thick, breadth and thickness measuring 2 cm. and 1 cm. respectively, just above the postzygapophyses. The fourth and fifth vertebrae are slightly smaller than the third, but do not show any remarkable features. The other remains belong to three smaller vertebrae, the smallest of which has a length of 5-5 cm., while its articulation surfaces have a height of 3-8 cm. The lower ends of both articulation surfaces are strongly recurved. The lower side of the centrum is grooved (PI. XVI, fig. 1 c). The proximal end of a haemapo- physis is preserved. The breadth of its upper end is 3-5 cm. and the length of its articulation surface is 2-7 cm. This must be one of the most anterior haemapophyses, for it fits exactly on to one of the large centra. Humerus. The distal end of the left humerus is preserved in a fragmentary condition, but the bone has apparently not suffered from pressure. The bone is broken off beneath the processus lateralis (PI. XIV, fig. 6 and text-figs. 8 and 9). The lateral side of the shaft is also lost. The proximal plate must have made an angle of nearly 90° with the plane of the distal end. A narrow, high, rounded ridge runs from the extreme end of the condylus lateralis upwards to the lower Annals of the Transvaal Museum 105 end of the crista radialis. The broad ridge running upwards from the condylus medialis seems to reach the medial side of the processus lateralis. The anterior surface between these two ridges is strongly concave ; in its distal part this concavity contains a circular depression. The anterior surface of the distal end is concave. The breadth of the distal end is 9 cm. Thickness medial condyle 3-9 cm., breadth 5 cm. Thickness lateral condyle 4-3 cm., breadth Text-fig. 9. Dromicosaurus gra- cilis. Distal end of left hume- rus. Outline of distal surface seen from below. Medial side to the left. Less than x 2*9 cm. The articulation surface of the lateral condyle is divided into two facets, an outer, which is directed downwards and inwards, and an inner, which is directed downwards and outwards, meeting in a ridge, which runs through the middle of the distal surface of the condyle and parallel to its lateral side. The lateral condyle leans over to the lateral side and the medial surface of the medial condyle is also directed forwards and sidewards. The distal surface of the medial condyle is convex; its medial portion is weathered. Radius. The proximal end of the left radius is present (PI. XIII, figs. 5 and 6). The length of the articulation surface is 4-7 cm. and its breadth is 2-7 cm. The surface is cylindrically concave, and the axis of this cylinder is directed from the anterior end of the antero-medial border to the posterior end of the postero-lateral border. The concavity is deepest at its anterior end, where it makes the deepest notch in the border. The posterior part of the antero- medial border and the anterior part of the postero-lateral border are both drawn upwards; the former is slightly higher than the latter. The edges of the articulation surface are broad and rounded. Pubis. When the fossil was discovered the pubes were still in situ , but all the covering matrix had completely weathered away. The proximal end of the right pubis was lost. The upper inner corner of the plate of the left pubis was present, but the portion which is bent downwards at this corner was absent. As preserved, the foramen obturatorium was therefore not completely closed. The anterior corner of the proximal end was weathered. The remainder io6 Annals of the Transvaal Museum of both bones, right up to their distal ends, was cracked, but otherwise perfect. The distal two-thirds of the bones showed a remarkable feature. The two plate-like portions of the pubes were coalesced along the middle and formed only one bone. No division between the two pubic plates could be found. Through the cracks it was apparent that the pubic plate was very thin along its middle line. During excavation everything was done to keep the pubes intact, and they were even got out on a lump of rock, which was placed in a box. While I was busy with the excavation of the further remains, somebody else was kind enough to meddle with the transport of this valuable specimen. It was placed on the head of a Kaffir-boy who took it down hill and overturned the box into a wheelbarrow before it could be prevented. After the pieces had been fitted together as far as possible the two pubes appeared as figured on PI. XVI, figs. 4 and 5. The length of the most com- plete of the two, the left pubis, cannot be accurately given, for the distal end cannot be fitted on to the remainder. The length of the bone as preserved, without the distal end, is 34 cm. Then comes a gap, which may have a length of 5 mm. medially and of about 2-5 cm. laterally. The distal end has a medial length of 6-5 cm. and a lateral length of 8 cm. The total length of the bone may therefore have been 42 cm. It is very improbable that the length of the gap was greater, for I am under the impression that the two pieces of bone were separated medially by a crack filled up with matrix. Laterally of course a piece of bone has been lost. The distal ends of both pubes show their complete breadth. If they are placed alongside of each other the maximum breadth of the pubic plate of the combined bones Text-fig. 10. Dromicosaurus gracilis. Outline at the distal end would be 16-5 cm. of section through middle portion of left The breadth of the plate would publS seen the distal end. Nat. size, have been less in the middle, because the distal ends have lateral projections. The pubic plate is thickest near its lateral side. The lateral border is fairly sharp, but the bone thickens rapidly from this border inwards. It attains its maximum thickness before the middle of the bone is reached and it then thins out gradually towards the medial line, where the extremely thin bone was coalesced with its fellow (text-fig. 10). The maximum thickness of the left half of the pubic plate above its middle is 2-5 cm. Near its distal end the thickness is less than 2 cm. A piece of the present medial edge of the bone has a thickness of 3 mm. The distal end is very much thickened, that is to say, the hinder surface of the distal end bulges considerably back- wards. The anterior surface of this end is slightly concave, through a slight thickening of the distal border. The lateral side bends slightly outwards at the distal end. The distal surface is very broad, because of the thick end (text-figs. 11 and 12). The hinder surface is concave from above downwards immediately above the distal surface. The maximum thickness of the distal end which lies midway between the sides is 3-5 cm. The maximum breadth of the distal surface is nearly 5 cm. The distal surface is convex and its lateral -hinder-medial border is half a circle. The anterior border is straight. Text-fig. 11. Dromicosaurus gracilis. Outline of lon- gitudinal section through distal end of left pubis. The anterior border is below, x Annals of the Transvaal Museum 107 This half circle and the front border lie in a plane which makes an acute angle with the anterior surface of the pubic plate. Nothing is present of the bent-down portion at the inner proximal corner of the pubic plate. A broken edge indicates how far this bent-down portion continued towards the proximal end. From here the pubic neck continues towards thearticulatio ileo-pubica. The section of the neck is triangular as shown in text-fig. 2. The lateral border of the neck is a continuation of the lateral border of the plate, it is only slightly concave forwards. The hinder-outer surface of the neck is slightly convex. The neck has an anterior surface and an inner surface, which stand at right angles to Text-fig. 12. Dvomicosaurus each other. The anterior surface rounds off gracilis. Outline of distal broadly into the inner surface. Both surfaces are surface of pubis seen from concave from above downwards. The distance between the lateral border and the inner posterior border of the neck is 4 cm . The upper anterior corner of the bone and the articulatio ileo-pubica are weathered away. The thickness of the head at the upper anterior end is about 5 cm. The processus subacetabularis is not only bent downwards at right angles to the neck, but its general direction is also practically perpen- dicular to the pubic plate. The medial border of its upper surface is fairly straight. The proximal portion of this upper surface has a high boss on its lateral border. The distal portion shows a deep concavity, which does not affect the medial border, but which continues over the lateral border. The medial surface of the processus is concave in all directions and strongly so from above downwards. Its general section is triangular. The length of the processus is about 7-5 cm. The breadth of its upper surface is less than 4-5 cm. The distal end of the processus subacetabularis is fiat and has a process which is directed downwards, forwards and inwards. This process stands practically at right angles to the processus subacetabularis. Distally it turns more and more inwards; its section is triangular. The lower end of the hinder surface of this process is damaged. The upper border of the triangular articulatio ischio-pubica, which is covered with small knobs and pits, has a breadth of about 4-5 cm. Its height as far as preserved is about the same. The breadth of the foramen obturatorium is 17 mm. Its length was more than 3 cm. Ischium. Both ischia are present, but they are very much broken. The left ischium is fairly complete, but it is broken in two and the parts are joined by a thick band of matrix. The thin upper anterior portion is badly broken and the extreme distal end is lost. The distal end of the right ischium is complete, but its proximal end is lost (PI. XV, fig. 4). If the bones are combined, it will be found that the length of the ischium is 34 cm. The distal portion is straight and the proximal plate is turned slightly outwards. The breadth of the proximal plate with the processus subacetabularis (PL XV, fig. 6) is 11 cm. A small piece of the posterior portion of the proximal end is broken off and the bone has suffered somewhat from pressure. The original total breadth may therefore have been 12 cm. The posterior portion of the proximal surface has a breadth of 6*5 cm. as pre- served. However, from this must be deducted 1 cm. for a seam of matrix, which passes in a sagittal direction through the posterior portion of the upper io8 Annals of the Transvaal Museum end. of the bone. The original breadth was, therefore, 5-5 cm. The surface is covered with numerous small knobs and pits. The outer border of this portion is formed by a long posterior and a short anterior sharp edge, which stand nearly at right angles to each other. The vertex of this angle is high and projects to the lateral side. The medial border of this surface passes upwards and forwards and, after reaching a point opposite the angle on the lateral border, downwards and forwards. There is a high boss on the proximal surface at the angle between these two medial borders. The medial and lateral angles of the border lie exactly opposite each other. The surface posterior to these two angles serves for articulation with the processus postacetabularis ilei. In gracilis. Left ischium. Outline of thearticulatio ischio-pubica seen from in front. Nat. size. Text-fig. 14. Dromicosaurus gra- cilis. Right ischium. Outline of distal surface seen from the distal end . The medial border to the left. Nat. size. 1 I front of the medial boss is a small triangular surface, which is bordered in front by a ridge ; this ridge is a continuation of the anterior part of the lateral border and runs obliquely across the proximal surface, meeting the medial border further downwards than its starting point. Distally to this ridge is the processus subacetabularis ischii. The medial border of this processus is extremely thin and slightly concave upwards. The upper surface of the pro- cessus is concave. The edge with the lateral surface is rounded. The lateral surface of the processus is convex from above downwards. The medial surface of the whole proximal end is strongly concave from above downwards. The distal end of the upper surface of the processus makes an acute angle with the articulation surface for the pubis. The articulatio ischio-pubica has a triangular shape (text-fig. 13) and the surface is covered with small knobs Annals of the Transvaal Museum 109 and pits. Its upper border has a length of 3-5 cm.; the height of the triangle is about 5 cm. The articulatio pubica makes an angle of about 6o° with the articulatio iliaca and an angle of about 40 0 with the general long axis of the ischium. The articulatio iliaca makes an angle of less than 25 0 with the general long axis of the bone. The thin anterior portion of the plate is damaged and its edge is broken off. Probably this edge was fairly straight or slightly convex between the lower end of the processus subacetabularis and the place where the proximal and the distal portions of the ischium form a curve. The thin anterior portion terminates at a distance of about 12 cm. below the proximal end. The posterior portion of the plate is thicker than the anterior portion. A few centimetres below the articulation surface its thickness is 18 mm. A broad groove starts at a short distance (1*5 cm.) below the articulation surface on the posterior edge of the lateral side. Through the twisting of the distal portion of the bone the groove soon passes on to its hinder surface. It cannot be made out how far this groove runs downwards. The distal end of the ischium has a flat hinder surface. Its section is triangular (text-fig. 14). The medial sides of the two distal ends lie against each other and although they are proximally separated by matrix, the two bones seem to be coalesced dis- tally. The specimen is slightly pressed sideways, but probably the two hinder surfaces lie in a plane. The two lateral surfaces meet in a ridge. The distal end is thickened. The border of the distal surface projects posteriorly, laterally and anteriorly (PI. XV, fig. 5). The distal surface is probably convex. The breadth of the hinder surface near the middle is 3-5 cm. and the thickness at the same spot 3 cm. The greatest breadth of the distal end of the right ischium is more than 4 cm. Its thickness is nearly 9 cm. Femur. The left femur is preserved (PI. XIII, figs. 2 — 4, and text-fig. 15), but in a rather weathered condition. The caput femoris is broken off and in its lower half the femur was badly broken in two places. The pieces were fitted to- gether by myself and joined with plaster of Paris. The length of the bone has not been influenced by these joints. It would not be exact to call the curvature of the diaphysis sigmoidal. Its middle portion is certainly convex forwards, but its upper end is not concave forwards as is usual, only less convex than the middle portion. The length of the bone is 49*5 cm. The breadth of the proximal end cannot, of course, be given. The preserved portion of the proximal end rounds off into the lateral side; from opposite the upper end of the trochanter major the lateral side bends more and more inwards till near the proximal end, where the convexity becomes greater to pass over into the proximal surface. The trochanter minor was on the piece which is broken off and lost. The trochanter major is broken off. The upper end of the broken surface is situated at a distance of 7-5 cm. from the proximal end of the bone, and the lower end of the trochanter at a distance of 13-5 cm. from the same end. Between the lateral side of the preserved portion of the trochanter and the general surface of the bone there is a broad groove. The femur is broad and thin between its proximal end and the trochanter major; further downwards it becomes thicker. A broad ridge starts at the fractured edge of the proximal end, and coming from the direction of the caput femoris runs towards the trochanter major. In passing the trochanter it turns in the direction of the condylus medialis. The highest point of this ridge lies immediately below the trochanter major; its height diminishes towards the proximal end and towards the distal end, in the latter case to such an extent that it loses itself on the medial side 8 no Annals of the Transvaal Museum below three-fifths of the length of the bone. The trochanter major does not lie on the top of the ridge, but is situated on its lateral side. Immediately below and medial to the lower end of the trochanter major the top of the ridge shows a broad, flat, rough surface. Lower down the top of the ridge is rounded and rough, the roughness narrowing down towards the middle of the bone, where it disappears. The surface of the bone enclosed between the medial side and the ridge, as far down as the upper end of the trochanter quartus, is concave. The lateral side of the anterior surface of the bone, below the lower end of the trochanter major and exactly opposite the whole length of the trochanter quartus, is slightly concave. A large oval foramen nutritivum, length 8 mm., breadth 4 mm., is situated on the lateral side of the top of the ridge and still within the rough surface, at a distance of 11 cm. from the upper end of the trochanter major. The medial side of the proximal end is hollow. This concavity, with the curve in the ridge and the convexity of the lateral side, marks the inward bend of the proximal end of the femur. The hinder surface of the proximal end is flat. Opposite the upper end of the trochanter quartus it rounds off over the lateral side of the bone into the anterior surface. Higher up, however, from opposite the middle of the trochanter major up to the proximal end of the lateral side, it rounds off into an elongated, flattish, lateral surface. This lateral surface is separated from the anterior surface by a heightened narrow ridge. The trochanter quartus begins at a distance of about 14-5 cm. from the proximal end of the bone. It terminates at a distance of about 21-5 cm. from that end. The height of the trochanter, as preserved, is 3 cm. Its upper edge is broken off and at the most it could have been 0-5 cm. higher. The medial side of the trochanter is overhanging; the lower end of the lateral side is steep, but the slope of its upper end is much less. These two portions of the lateral side meet in a clearly visible rounded ridge, which runs in the direction of the upper outer cor- ner of the proximal end and seems Text-fig. 15. Dromicosaurus graphs. Left ^ femur. Outline of distal end seen from below. Medial side to the right. to continue further upwards than the much higher medial ridge of the trochanter. The whole trochanter is situated nearer to the medial than to the lateral side of the bone ; moreover, the upper end lies nearer to the medial side than the lower end and therefore the edge of the trochanter lies in the direction of the condylus lateralis. Medially to the trochanter quartus there is a large, flat, rough surface, which has apparently served for the attachment of muscles. It looks mostly inwards and only slightly backwards. A ridge starts at the upper end of this surface, and runs from the lateral side on to the posterior surface in the direction of the lateral corner of the proximal end, passing well above the upper end of the trochanter quartus, where it terminates. The breadth of the diaphysis over the trochanter quartus is 6-5 cm. Its thickness, measured from the top of the lower end of the trochanter to the top of the ridge on the anterior surface, just above the foramen nutritivum, is 8*5 cm. At about 18 cm. above the distal end these dimensions are 7-5 cm. and 3-5 cm. respectively. The thickness of the distal end, measured from the anterior Annals of the Transvaal Museum hi surface to the lower end of the fossa intercondyloidea, is 3-5 cm. Its thickness over the condylus medialis is 8-i cm. and over the condylus lateralis 9 cm. The upper end of the fossa intercondyloidea is damaged. It starts at least 12 cm. above the distal end. The condylus medialis is damaged; its breadth as preserved is 5 cm. Its height is about 6*5 cm. The condylus lateralis is much narrower; its breadth is 3*3 cm. The breadth of the distal end (text- fig. 15) is about 11*5 cm. The distal articulation surface is full of cracks and could not be well cleaned of matrix. The lateral side of the distal end is hollow and the hinder end of the condylus lateralis stands, out slightly to the lateral side. The anterior surface of the distal end is hollow along its middle. Seen from the front it seems as if the lateral side of the bone runs slightly further downwards than the medial side. The diaphysis is hollow. The bone only forms a thin shell round a very large cavity. Tibia. Only the right tibia is preserved and that in perfect condition. It has a large proximal articulation surface and a straight shaft (PI. XIV, figs. 1 — 3, and text-figs. 16 and 17). During its excavation, the bone of the diaphysis parted in small fragments from the filling of its cavity; however, all the pieces were fitted and glued together on the spot. The total length of the bone is 3 7 • 5 cm . The head of the tibia (text-fig. 16) has a length of 13 cm., measured from the hinder end of the medial condylus to the edge of the anterior crest. Its greatest breadth, transverse to this greatest length, is 8-5 cm. The length of the lateral side of the head is 9 cm. Anteriorly the head of the tibia has a laterally directed process, the tuberositas Text-fig. 16. Dromicosaurus gracilis. Right tibiae. The lateral border of the tibia. Outline of proximal surface seen proximal articulation surface is di- above- Lateral side to the right, rected straight backwards from the tuberositas tibiae, but before reaching the lateral condylus it becomes slightly concave; the remainder of this border is convex. The medial border runs from the tuberositas tibiae inwards and curves slightly backwards until at about 3-5 cm. from the anterior end the convexity becomes greater. It diminishes again at about 5 cm. from the anterior end; from this point further backwards the medial border is only slightly curved, but the convexity is greater around the condylus medialis. The posterior border is slightly convex behind each condylus and notched in the middle. The proximal articulation surface shows a large shallow concavity above the medial condylus ; another large concavity is situated on the anterior end, while the upper surface of the lateral condylus is convex. The lateral posterior side of the surface is slightly lower than the medial posterior side. From the hinder border the articulation surface goes steeply upwards and forwards. The hinder border is overhanging; it rounds off abruptly into the lateral border; its medial end is slightly damaged. The shaft narrows down quickly immediately below the head. The distal end is transversely broad. The medial one-third of the distal articulation surface is flat and slopes slightly from its anterior edge backwards 8—2 112 Annals of the Transvaal Museum and downwards. The lateral part of the distal articulation surface consists of an anterior and a posterior portion (text-fig. 17). The anterior portion, which is the distal surface of the processus tibiae anterior distalis, occupies a much higher position than the posterior portion, which is the distal surface of the processus tibiae posterior distalis. The distal surface of the processus posterior is continuous with the medial portion of the articulation surface. It slopes upwards from the lateral to the medial side and from the posterior towards the anterior border. The medial portion of the articulation surface is con- nected with the distal surface of the processus anterior by means of a short and very steep surface. The distal face of the processus anterior slopes from its medial end upwards and outwards, and from its anterior border backwards and slightly upwards. Medially the processus posterior is nearly as broad as the processus anterior. Towards the lateral side it becomes narrower and it terminates in a point. This end is bent very slightly forwards. Both ends of the distal surface of the processus anterior are broad, but the inner end is slightly broader. Apparently, the lateral end of the processus anterior does not project further outwards than the processus posterior. The highest point of the lower surface of the processus anterior is situated 37 mm. above the lowest part of the processus posterior. There is a slightly concave surface (sulcus malleoli tibiae) imme- diately behind the lateral end of the processus anterior. This surface passes downwards be- tween the processi and is continuous with the concave upper anterior surface of the processus posterior ; it passes inwards for a short distance only. The breadth of the distal end, along the front border of the articulation surface, is 82 mm. This front border is laterally slightly concave and medially, where a small piece of the edge is broken off, it is slightly convex. The medial border makes an acute angle with the anterior border and an obtuse angle with the posterior border. The borders round off into each other. The breadth of the medial side, measured parallel to the medial border, is 56 mm. The posterior border converges slightly with the anterior border towards the lateral side; its length is 60 mm. The distance from the anterior medial corner of the distal end to the lateral end of the processus posterior is 1 mm. longer than to that of the processus anterior. A broad high ridge runs down from the tuberositas tibiae in the direction of the anterior medial corner of the distal end. It does not reach this corner, however, for it terminates at about 6 cm. above the distal end. The surface between the lower end of this ridge and the distal end of the bone is concave. The lateral anterior edge of the distal end is broadly rounded below. Higher up the edge becomes a fairly sharp ridge ; then again the ridge becomes broad and low and finally disappears above the middle of the bone. There is a broad, oval and very low boss on the lateral side of the bone, below the notch in the lateral border of the proximal surface; its upper end lies about 4 cm. below the notch and its lower end about 8 cm. Its breadth is about 2-5 cm. The thickness of the bone in the middle is 5-5 cm. and its breadth at this place about 4 cm. The thickness of the distal end is about 5 cm. Text-fig. 17. Dromicosaurus gra- cilis. Right tibia. Outline of distal surface seen from the distal end. Anterior border below. Lateral side to the right, x T Annals of the Transvaal Museum 113 Fibula. Of the fibulae only the right bone has been found and this is in excellent condition. It has a length of 34*5 cm. v. Huene has already pointed out that one can readily ascertain whether the fibula belongs to the left or to the right side. The lateral side of the proximal end is convex. The thin and low portion of the proximal end is anterior, the thick portion posterior. In the following description the bone is so placed that the inner border of the posterior portion of the proximal articulation surface is directed straight backwards (PI. XV, figs. 1 — 3). The proximal articulation surface is convex from side to side and its anterior portion is bent inwards (text-fig. 18). The whole surface is covered with irregular grooves and ridges. The anterior portion of the proximal end of the bone is strongly directed inwards; it is really a high, thin, anterior process. Its most anterior point lies far below the articulation surface. The lateral surface of the posterior portion of the proximal end is evenly convex. That of the anterior portion is slightly concave. Between these two portions the lateral surface is strongly convex. The posterior portion of the medial surface is flat; its anterior portion is slightly concave. There is a broad low boss slightly below the middle of the inner surface of the proximal end. From the posterior corner of the proximal end a broad ridge passes downwards and inwards for a short distance; it narrows down quickly and ends opposite the middle of the broad boss on the inner surface. The greatest breadth of the proximal end is 7-7 cm. Its greatest thickness is 2-8 cm. The thickness of the bone over the broad boss is 3 cm. The shaft narrows down rapidly immediately below the head. At a dis- tance of 9 cm. from the proximal end the, breadth of the bone has diminished to 3*5 cm. and the thickness to 2-1 cm. At a distance of 15 cm. from the proximal end the breadth of the bone is reduced to 3 cm., but the thickness has in- creased and is at this spot 3 cm. The greatest horizontal dimension of the shaft at this distance from the proximal end is 3*4 cm. This is due to a broad, high ridge on the lateral surface of the bone. This ridge starts very gradually at a distance of about 9-5 cm. from the proximal end on the anterior border of the lateral side of the bone. It ends at about 19 cm. from the proximal end. The distal end of the ridge is situated in the middle of the lateral surface of the shaft. The posterior side of the ridge is slightly concave at its upper end, but at its lower end it passes gradually into the posterior surface of the bone. The anterior side of the ridge is slightly concave. The anterior edge of the bone passes down as a low ridge along the medial side of this concave surface. There is a large oval depression on the anterior surface of the bone, medial to the low ridge just mentioned and looking slightly inwards. The medial side of this depression is slightly elevated above the medial surface. The upper and lower ends of the depression lie at a distance of 12 cm. and 15 cm. respectively from the proximal end. Slightly below the oval depression the low ridge subsides into the general surface of the bone. At a short distance from the depression the anterior edge again assumes a ridgy appearance. This ridge, however, is not a continuation of the preceding ridge, for the lower end of the latter lies higher than and lateral to the upper end of the former. This ridge runs downwards for a short distance and then also subsides into the general surface of the bone. A foramen nutritivum of about 6 mm. length micosauvus gra- cilis. Right fibula. Outline of proximal end seen from above. Medial side to the left, x A. Annals of the Transvaal Museum 114 and 2 mm. breadth is situated on the medial surface, at a short distance below the oval depression. The hinder surface of the bone is narrow and rounded. The edge between the hinder surface and the medial surface is rounded in its upper two-thirds. At the upper end of its lower third it forms a short, sharp and low ridge, the lower end of which lies 10 cm. above the distal articulation surface. From this ridge two diverging ridges run towards the distal end. The posterior one is low, broad and rounded at its upper end ; at its lower end it is a high and broad ridge, the edge of which, however, is broken off. As the medial-posterior side of the distal end is weathered, one cannot be sure whether this ridge reached the border of the articulation surface ; there are indications that it did not. The anterior ridge is also low, broad and rounded at its upper end, but its lower end is high and narrow, this being the result of the medial and anterior surfaces meeting at an acute angle. This ridge runs right down to the antero-medial corner of the distal end. The medial surface between the two ridges is concave. The posterior and anterior surfaces of the distal end are narrow and slightly convex in horizontal section. In a vertical direction they are concave, especially the anterior surface. The lateral surface of the distal end is narrow and very convex in horizontal section ; it is slightly concave from above downwards. The distal articulation surface (text-fig. 19) has somewhat the shape of an oval. Its long axis makes an angle of slightly more than 450 with the sagittal line. Its anterior end lies medially and its posterior end laterally. The articulation surface is covered with irregularities but is generally flattish. It slopes from the antero-medial end downwards to the postero-lateral end. The difference in height of the two corners is about 16 mm. The postero-medial border is weathered, but was probably an evenly convex line from the antero-medial end to the postero-lateral end. The border of the postero-lateral end is broadly rounded, that of the antero-medial end bluntly pointed. There is a short, straight anterior border running from the antero-medial end outwards till past the middle of the distal end. The front border is connected by another straight border with the postero-lateral end. The distance from the antero-medial corner to the postero-lateral corner is 6-2 cm. The breadth of the distal end as preserved is 3*3 cm. The diaphysis is bent and concave medially. At a distance of 20 cm. from the proximal end the breadth of the shaft (transverse dimension) is 2-5 cm. and the thickness 2-6 cm. Foot. The foot is represented by some well-preserved fragments. Metatarsale I is represented by two pieces, one of which is the proximal end of the right bone (text-fig. 20). Its breadth is 5 cm. and its thickness 2*4 cm. The length of the piece is 5-5 cm. There is a narrow ridge in front which passes into the rounded anterior surface of the shaft before reaching the end of the piece. A broad ridge runs downwards from the middle of the lateral border of the proximal articulation surface; it reaches the anterior border of the shaft at the fractured end. Posteriorly the bone has a broad rough edge. Further downwards this edge is smooth and passes into the rounded posterior surface of the shaft. There is a slight angle between the rough and the smooth surface. The medial surface bends to the lateral side near the posterior border. The proximal surface is hollow in the middle; Text-fig. 19. Dro- micosaurus gra- cilis. Rightfibula. Outline of distal end seen from below. Anterior border above. Medial side to the right, -x Annals of the Transvaal Museum ii5 this cavity passes forwards and outwards and over the anterior lateral border. The other piece is.a fragment of the distal end of the right bone (PL XIII, fig. 8). The globose part of the articulation surface has a thickness of 2-7 cm. The lateral collateral pit has a tremendous size. It has a length of 2 cm. and a breadth of 1-2 cm. Its upper anterior and lower posterior ends are open. The cavity has a more or less cylindrical shape. That this piece must belong to the right bone is shown by the fact that its latero- posterior surface is concave close to the medial fracture, which concavity is a portion of the large concavity on the latero-posterior surface, well-known from other first phalanges. The antero-medial surface of the bone is separated by a ridge from the articulation surface. Metatarsale II is only represented by the proximal end of the right bone (text-fig. 20). The proximal articulation surface has the shape of a quadrangle. All four sides of this quadrangle are concave. The medial side is deepest in the middle, while the deepest place of the lateral side lies nearer to the hinder border. The lengths of the different borders of the proximal surface are: medial 5-7 cm., lateral approximately 6 cm., posterior 4-1 cm. and anterior approximately 3 cm. Ridges run from the corners of the proximal surface downwards on to the shaft. The upper end of the antero-lateral ridge is weathered away, but very probably it was sharp. The antero-medial ridge is rounded. The base of the postero-medial ridge is broader than that of the postero-lateral one. The proximal surface of this last ridge is slightly lower than that of the former one. The postero-lateral ridge stands further out from the bone than the postero-medial one. The anterior border of the proximal surface overhangs the anterior surface of the bone. The length of the piece is 4 cm. Metatarsale III. There is one piece which, I take it, belongs here, viz. the distal end of the right bone. Its medial side is missing (PI. XIV, fig. 4). The lateral side has a thickness of 2-4 cm. The lateral collateral pit is very deep. Its hinder edge has a small process behind the middle of the pit and its anterior edge has two knob-like processes with a groove in between. The upper knob-like process is separated from the process on the hinder edge by a deep groove. A rounded ridge passes on the anterior surface inwards and slightly upwards from the upper knob-like process. The anterior surface is slightly concave between this ridge and the articulation surface. I am not quite certain of the identification of this bone. The last bone of this collection is also difficult to identify. It is probably the proximal end of the left ulna and although there is no certainty on this point I will here describe it as such. The length of the piece (PI. XIV, fig. 5) is nearly 8 cm. Its antero-medial side is weathered. As preserved, the length of the lateral border of the proximal surface (text-fig. 21) is 3-6 cm. The length of its posterior border is 5 cm. The length of its antero-medial border was greater than 5-6 cm. The proximal surface is divided into two portions by a high ridge, which passes over the surface from the middle of the posterior to the middle of the antero-medial side. The portion of the proximal surface which lies medial and posterior to this ridge is convex and slopes down towards Text-fig. 20. Dr omicosaurus gra- cilis. Metatarsale I and II of the right side. Outline of proximal end of the bones, seen from the proximal side. x i- ri6 Annals of the Transvaal Museum the postero-medial corner of the surface. The portion of the proximal surface which lies lateral and anterior to the ridge slopes more strongly forwards and is concave medially. The posterior border of the surface is concave laterally and convex medially. The lateral border overhangs the lateral surface of the shaft. The antero-medial corner is slightly lower than the postero-lateral one. Text-fig. 2 1 . Bromicosaurus gracilis . Outline of the proximal end of the left ulna(P), seen from the proximal side. Posterior side to the left and lateral side above. x i- If this is actually the proximal end of the left ulna, then a large portion of the antero-medial angle of the head has worn away. Discussion. It will not be necessary to compare with the Plateosaurid Eucnemesaurus , for the present form is plainly an Anchisaurid. The trochanter quartus of our form lies relatively much deeper than that of Plateosaurus stormbevgensis. A comparison with Massospondylus carinatus is difficult, because most of the type remains are fragments. The head of the pubis shows some important differences. In the first place the length of the head of the pubis of Masso- spondylus .carinatus as figured by v. Huene (4, p. 38, fig. 51), and measured from the articulatio ischio-pubica straight across to the anterior surface, is 9-6 cm. In our form the upper anterior portion of the head is damaged. As preserved, the length of head and processus subacetabularis together is 7-1 cm. In the above the total length of the undamaged head was estimated at 7-5 cm. It can certainly not have been more than 8 cm. It appears, therefore, that the pubic head of Massospondylus carinatus is longer than that of our form. It is, however, apparent from the femur that our form is a bigger animal than the type of Massospondylus carinatus. The pubic head is, therefore, relatively much longer than that of our form. The thickness of the head is about half that of our form. The section of the neck of the pubis, which has the shape of a very elongated oval with pointed ends, differs considerably from that of the present form. The coalesced portion of the ischia shows different sections to that of the form just described. Sections of the coalesced portion in our form are tri- angular everywhere, they do not even approach the rounded form of the distal end of the Massospondylus fragment. The length of the femur of Massospondylus carinatus is estimated at 43 cm. The length of the lower portion below the trochanter quartus is about 22 cm. In our form these lengths are 49-5 cm. and 28 cm. Our femur, therefore, is longer. In the femur of Massospondylus carinatus, however, the trochanter major lies 9 cm. below the proximal end, while in our form this distance is only 7-5 cm. The breadth of our femur above the trochanter major is 7*5 cm. and its thickness there is 3-6 cm. (the bone is not crushed). In Massospondylus carinatus these dimensions are 7 cm. and 4-4 cm. respectively. These dimen- sions show, therefore, different relations in the two bones. The tibia of Massospondylus carinatus is not complete. However, the proximal and distal ends are preserved, v. Huene gives the following measure- Annals of the Transvaal Museum 117 ments of the head: length of the medial side n cm., of the lateral side 9-5 cm. and greatest breadth 7-5 cm. (4, p. 42). These dimensions in our form are 13 cm., 9 cm. and 8*5 cm. respectively. From these measurements it will be clear that the relations of the two bones are different. In Massospondylus carinatus the distal end of the tibia has the following dimensions: length anterior border 6 cm., medial border 4-5 cm., posterior border less than 4 cm. and lateral border 4 cm. In our form these dimensions are respectively 8*2 cm. , 5-6 cm., 6 cm. and 4*2 cm. In Massospondylus carinatus the medial border of the distal end is longer than the posterior border, in our form it is the reverse. The above-mentioned differences may be regarded as sufficient to exclude our form from the genus Massospondylus. Massospondylus Harriesi is very much smaller than our form. The distal end of its femur, measured from the lower end of the trochanter quartus, is 15*5 cm. In our form this portion measures 28 cm. The proximal end of the tibia of Massospondylus Harriesi measures 10-2 cm. x 5-2 cm. If the same relations existed in our form as in Massospondylus Harriesi the tibial head of the last should have a length of 13 x 15-5 : 28 = 7-2 cm. The head of the tibia of Massospondylus Harriesi is therefore much longer than in our form. Length and breadth of the head of the tibia in our form are 13 cm. and 8-5 cm. respectively. If the same relations existed, the breadth of the head of the tibia in Massospondylus Harriesi, with regard to its length, should be 8-5 x 10-2 : 13 = 6-67 cm. Therefore, the head of the tibia in Massospondylus Harriesi is relatively narrower than in our form. A comparison with Aetonyx palustris is difficult, because what is present in the one is missing in the other (7, p. 304). In Aetonyx palustris the width of the lower end of the humerus is 5-7 cm. In our form it is 9 cm. The length of the humerus in Aetonyx is 17-4 cm. If the same relations exist in our form, the length of our humerus should be 17-4x9: 5-7 = 27-5 cm. In Aetonyx palustris the lower end of the delto-pectoral ridge lies 10 -i cm. from the upper end of the bone. This measurement and the length of the bone have been verified in the figure and were found to be correct. The distance of the lower end of the delto-pectoral ridge from the distal end of the bone, measured in the figure, is 9-2 cm. This may be slightly more in reality through fore- shortening, and also because of the damaged condyle. With the same rela- tions the lower end of the delto-pectoral ridge in our form -should lie at a distance of 9 x 9-2 : 5-7 =^i4-4 cm. from the distal end. With regard to the curvature at the upper end of our fragment and after comparison with other humeri, I come to the conclusion that it must lie at a distance of, at the very most, 13 cm. from the distal end. However, with a slightly longer upper end the total length of the bone would then come near the result obtained above. In Aetonyx palustris the proximal width of the second metatarsale is 2-5 cm. and that of the third metatarsale 2-6 cm. It could not be made out whether Dr Broom means the greatest dimension of the proximal ends or the breadth along one of the borders of the bones. From the description of Massospondylus Harriesi, however, I conclude that the greatest dimension is meant. In our form the greatest dimension of these two ends is 6-7 cm. and 5-6 cm. respect- ively. The greatest breadth of our metatarsale II is 4-1 cm. and of metatar- sale III 3-6 cm. In both cases, therefore, the proximal end of our metatar- sale II is larger than that of metatarsale III, whereas in Aetonyx palustris the proximal end of metatarsale III is only slightly wider than that of meta- tarsale II. The head of the tibia is relatively shorter in Thecodontosaurus skirtopodus than in our form (4, p. 44). The femora of Thecodontosaurus Browni (2, p. 124 n8 Annals of the Transvaal Museum and 4, p. 46) are much more curved than in our form. In Thecodontosaurus Browni the femur has a length of 24 cm. and the lower end of the trochanter quartus lies 11 cm. below the proximal end. In our form these measurements are 49*5 cm. and 21-5 cm. respectively, and it easily follows that the trochanter quartus lies relatively higher in our form than in Thecodontosaurus Browni „ The trochanter major lies also relatively higher in our form. Although our tibia is shorter than that of Gryponyx africanus, its head is absolutely broader. The relations of the head of the tibia in Gryponyx afri- canus are altogether different from those in our form. The shape of pubis and ischium is also quite different. A comparison of the ischium of the present form with the well-preserved one of Teratosaurus suevicus H. v. Meyer (5) shows that, where the last is twice as long as broad, the ischium of our form is more than three times as long as broad. Casually connected with this greater relative breadth is the fact that the articulation surfaces of the ischium of Teratosaurus suevicus make much greater angles with each other and with the shaft of the bone than in our form. In the present form the articulatio iliaca makes an angle of less than 25 0 with the long axis of the ischium. The articulatio pubica makes an angle of about 40° with the long axis of the ischium and of about 6o° with the articulatio iliaca. In Teratosaurus suevicus these different angles are respectively 550, 750 and 130°. If we compare the fibula of the present form with that of Plateosaurus Quenstedti, we are at once struck by the great difference between the two. The proximal end of our fibula has a large posterior process, while the pos- terior border of the upper end of the fibula of Plateosaurus Quenstedti is evenly rounded. The shaft in our form becomes a narrow rod in the middle, while in the fibula of Plateosaurus Quenstedti it is proximally very broad and distally only slightly narrower. The medial view of the distal end of the fibula of Plateosaurus Quenstedti does not show the medial ridge of our form. The anterior ridge of our form is broad distally, but it does not bear a smooth surface as in Plateosaurus Quenstedti. Our fibula also differs from Gresslyo- saurus cf. Plieningeri through its large proximal posterior elongation. It appears, therefore, that the present form is new and belongs to a new genus. The form is allied to Aetonyx and Massospondylus and is, therefore, an Anchisaurid. I propose to call this new form Dromicosaurus gracilis n. g. et sp. (8pofui<6<; = quickly walking). The slender leg must have enabled the animal to go quicker than, for example, Eucnemesaurus and also implies a more slender form. Massospondylus Browni Seeley. The remains which are being described under this name have been dis- covered on the farm St Fort (Letsoanastad No. 528 of 1905 map of Bethlehem District) by its owner, Mr H. Walker, who very generously presented them to our institution. They consist of a vertebral column, front and hind legs, pectoral and pelvic girdle, all of one animal. The exact spot of the occurrence is in the Red Beds on the boundary between Clifton and St Fort immediately to the north of an old road through a Nek crossing this boundary, the Nek lying to the north of a hill which projects into the big bend of the Little Caledon River on St Fort. The spot lies about ten feet below the thick banks of the Cave Sandstone. A good search was made for the skull and a portion of the neck, which could not have been removed by weathering, but nothing further was found. Annals of the Transvaal Museum 119 Cervical Vertebrae. Five of the neck vertebrae have' been preserved. They form an unbroken series, the last of which is at the same time the last neck vertebra (PI. XXI, fig. 3). They are all in very bad condition, flattened and crushed, and the last one is nearly unrecognisable. The lengths of their centra are, from the anterior one backwards, 9 cm., 9 cm., 7-5 cm., 7*5 cm. and ? cm. As preserved the first and second of the series are slightly keeled, especially anteriorly and posteriorly. The third and fourth are more strongly keeled, although their lower borders are still concave. The fifth is strongly keeled, but that is all that can be said about it. Most of the dimensions are useless, and the general appearance of the vertebrae will be best understood from the figures. The breadth of the dorsal spine in the first four vertebrae is respectively about 5 cm., about 6 cm., about 5 cm. and 4-5 cm. The praezygapophyses of the second vertebra have a length of about 4-5 cm. The articulation surfaces of all the zygapophyses con- verge downwards. There is a slight ridge on the neural arch, just above and parallel with the neural suture. Posteriorly this ridge is evenly rounded, but in the third vertebra its lower side is hollow over about two-thirds of its length starting in front, and it forms, therefore, a downward leaning crista over this distance. In the more anterior vertebrae this crista is shorter. The diapophysis is situated on the anterior portion of this crista. It is not pre- served. The parapophysis, which is situated near the anterior edge of the centrum, is either broken off or covered by the proximal end of a rib. Some of the ribs of the neck vertebrae are in good preservation. The tuberculum branches off from near the proximal end of the rib under an acute angle. Its articulating end is broken off in most instances ; a portion of it is preserved in one rib and it is shown to be hollow upwards. In this specimen its length is 2 cm. This rib probably belongs to the fourth vertebra of the series; its probable fellow is also present. The capitulum is an inward process, which is not given off at the extreme proximal end of the rib, but slightly distal to it. The rib therefore has a small anterior process in front of the base of the capitulum. The capitulum is a thick round process directed forwards and inwards, with an expanded, circular, hollow articulation surface. In the ribs which probably belong to the fourth vertebra of the series its length is 5 and 7 mm. and the breadth of its articulation surface 8 mm. The length of the capitulum of the third vertebra is 3 mm. and the breadth of its articulation surface 9 mm. In the second vertebra the rib has a capitulum with a cup- shaped articulation surface attached to its inner side and near its end. The ribs were so long when they were found that their thin distal ends projected beyond the hinder ends of the vertebrae. The right rib of the first vertebra still shows a length of 8-5 cm. and both ends are broken off. Dorsal Vertebrae. All the dorsal vetebrae have been preserved, but in such a bad condition that the exact number of dorsals cannot be given with absolute certainty. If the remains of a vertebra behind the last neck vertebra are those of the first dorsal, then the second dorsal is only represented by its zygapophyses. From the third onwards the series is complete. It is difficult to decide which vertebra is the last dorsal. If the above assumption is right, then the fifteenth of the series was the first between the ilea. For reasons which will be stated later, I take this vertebra to be the first sacral. The number of dorsals there- fore is 14, if the assumption of the first dorsal is right. It is not impossible, although very improbable, that a vertebra is missing between the last neck 120 Annals of the Transvaal Museum vertebra and what is here called the first dorsal. It is also possible that more than one vertebra is missing between the first and the third dorsal. This, however, is also very improbable, for the fossil was excavated by myself and there was sufficient space for one vertebra only at this spot. It is therefore fairly certain that the fossil did not have more than fourteen dorsal vertebrae. The postzygapophyses of the first dorsal are very long, their length being 2 cm. Those of the second and third are slightly shorter, while those of the fourth have a length of i*2 cm. The processus spinosus of the fourth vertebra has a height of 3 cm. while its length is slightly more than 3 cm. The length of the centrum of the fourth vertebra is about 5 cm. ; that of the sixth, the ninth, the tenth, the twelfth and the thirteenth is about the same. All the vertebrae are so flattened, however, that it is impossible to give accurate dimensions, and the above are only given to assist one in getting some idea of the size. The eighth vertebra shows a thick and nearly vertical posterior supporting ridge of the processus transversus. The ninth shows also the narrow anterior one, which slopes forwards and downwards; at its anterior end lies the large parapophysis. In the tenth vertebra (PI. XXI, fig. 2) the posterior supporting ridge is thick and connects the processus transversus with the posterior edge of the centrum. The anterior supporting ridge lies nearly horizontal. At its anterior end and at the base of the praezygapophysis lies the large parapophysis, completely above the centro-neural suture. In the eleventh vertebra the posterior supporting ridge is thick and the anterior ridge thin, but the last is shorter than in the tenth vertebra. In the twelfth vertebra the posterior supporting ridge is thick and the anterior one thin; the anterior one is shorter than in the eleventh dorsal and to such an extent that the transverse process and the parapophysis are nearly touching each other. The processus spinosus of the tenth vertebra has a height of 3-7 cm. and a breadth of 4*4 cm. Sacral Vertebrae. The three sacral vertebrae have also been preserved, but they are in the same condition as the dorsals. The second sacral is the longest; its length is 5-4 cm. That of the first sacral is 4-4 cm. and that of the third 3-8 cm. These three vertebrae were found between the ilea. In the Plateosauridae the second sacral is the longest and for this reason I take the longest of the present three to be the second sacral. The vertebra in front of it must then be the first sacral and the one anterior to that the last dorsal. Caudal Vertebrae. An unbroken series of eleven caudal vertebrae has been preserved (PI. XX, fig. 5). The first caudal is only represented by an anterior and a posterior piece of its centrum. Haemapophyses are present on all the caudals beginning with the first, except on the fourth from which it is missing. I cannot be quite sure that there is nothing missing between the last sacral and what I call the first caudal, because there was a slight fault in the rock which had displaced the two bones with regard to each other. I think it very unlikely, however, that a vertebra is missing between them. The first caudal was wedge-shaped. The length of the lower surface of the centra and their anterior height is as follows: second caudal 4 cm. and 4-5 cm., third 4 cm. and 4 cm., fifth 4-8 cm. and 3-7 cm., sixth 4-5 cm. and 3-4 cm., seventh 4-5 cm. and 3-4 cm., ninth 4*2? cm. and 2*8 cm., eleventh 4*3 cm. and 2-6 cm. Through the crushing of the vertebrae, which, however, is not so great as in the dorsals, these measurements do not give the actual size of the original uncrujshed bones; they are, however, not far out, and they give some idea of the relative Annals of the Transvaal Museum 121 size of the vertebrae. From the second caudal onwards the lower surface of all the centra is grooved. This groove is more clearly visible in the sixth to the eleventh caudal. The groove is deep in its posterior end, where it cuts slightly into the lower margin of the posterior articulation surface. The transverse processes are broad and thin; they are directed outwards, back- wards and upwards. Their base is still broad in the last vertebra. The dorsal spines are high and narrow; even the dorsal spine of the third caudal only attains a breadth of 1*7 cm. in its upper end, which is much less than half that of the tenth dorsal. Its length was about 6 cm. The dorsal spines of the sixth to the tenth caudal are preserved. There is an interval between them and the praezygapophyses. Their anterior edge is sharp. The length of the seventh is nearly 6 cm. ; that of the tenth nearly 4-5 cm. The posterior edge of the dorsal spine is a sharp ridge above, but below this sharp ridge runs into a groove formed by the backward projecting lower end of the sides of the spine. The praezygapophyses are long and stand out forwards and upwards. Their articulation surfaces are perpendicular in the sixth to the tenth caudal (the others are not or too badly preserved) and each is parallel with its fellow. The postzygapophyses are short and are situated high up behind the dorsal spine. The haemapophysis of the first caudal has lost its proximal and its distal end. As preserved its length is 9-5 cm. That of the second caudal has lost its distal end only and its length as preserved is 1 1 cm. There is a deep groove downwards from the proximal end on the anterior and on the posterior edge. The haemapophysis of the ninth caudal was also longer than 11 cm. The proximal end has two articulation surfaces, a larger anterior one and a narrow, half-moon shaped posterior one. The two surfaces stand in such a way that, when they are applied to the vertebra, the axis of the haemapophysis makes an angle of about 45 0 with the axis of the vertebra. Shoulder Girdle. Scapula and coracoid of the right side are present, but also in very bad condition. The scapula (PI. XVII, fig. 1) has a length of 25-5 cm. Its breadth at the narrowest place in the middle is 3-8 cm. Its breadth at the distal end cannot be given accurately for the upper corner is broken off ; as preserved it is 5-2 cm. ; it may originally have been 7 cm. The breadth at the proximal end, from the top of the processus deltoideus as preserved to the lower border of the facies glenoidalis pro humero, is 8-5 cm.; originally this may have been 10-5 cm. The thickness of the scapula at its distal end is 0-9 cm. Near the lower border of the curved portion the thickness is 1-7 cm. The thickness of the bone at the articulation surface for the humerus is 2-8 cm. The distal edge is broken off and nothing is visible of a thickening at this border. This may also be the result of crushing. The processus deltoideus is broken off. There is an indication of a circular depression in front of the processus deltoideus, but there is not a trace of a distal sharp edge to this depression. The articulation surface for the humerus is broad. Its inner portion projects further downwards and forwards than its outer portion. The distal end of the scapula is straight; in its proximal half, however, the bone is curved with the concavity inwards. The coracoid (PI. XVII, fig. 1) is an outwardly convex and inwardly concave plate. A large piece of its upper portion is lost. The upper portion is thin and the lower portion thicker. The articulation surface for the humerus is too much damaged for description and the articulation surface for the scapula is covered by that bone. The foramen supracoracoideum lies at a distance of 122 Annals of the Transvaal Museum f 5 -5 cm. above the lower border. It is a round opening of about 9 mm. dia- meter. The foramen passes obliquely through the bone, upwards, backwards and inwards. Humerus. Both humeri have been preserved, but both have suffered much from pressure. The proximal plate of the right humerus has been flattened and broadened, while the plate of the left one has been compressed. The shaft of the right humerus has been twisted. The length of the humerus is 20-5 cm. in the left (PI. XVII, figs. 2 and 3) and 21-5 cm. in the right bone. The broad surfaces of the distal end of the humerus make an angle with those of the proximal plate. The thick medial border of the proximal plate is concave. The caput humeri is not situated at the extreme end of the medial border, but it lies more inwards on the upper border, of which it forms the highest point. From here the upper border runs outwards and downwards to the lateral side. The fossa bicipitis lies below this border and is bounded medially by the thick medial edge and laterally by the convexity of the crista radialis. It does not pass on to the shaft. The crista radialis projects from the lower endTrf, the lateral side of the plate. Its height is about o-6 cm., and its length about 5 cm. The upper end of the crista stands out from the rest of the lateral border towards the medial side and consequently the medial surface of the crista and a portion of the plate near it is convex. The posterior surface of the plate is convex, except the portion medial to the caput humeri, which is concave, and the posterior surface of the crista, which is also concave. There is a circular depression on the posterior surface inwards of and slightly above the lower end of the crista. This cavity is not accidental, for it is present in the same position in both bones (fig. 2 of PI. XVII shows it plainly). The lower end of the crista radialis lies at a distance of 11 cm. from the distal end of the condylus lateralis. The distance between the upper end of the crista and the medial upper corner of the bone is 10 cm. in the left bone and is nearly 12*5 cm. in the right one. It seems as if the dimensions of the left bone are nearer to the original. The breadth of the shaft at its narrowest place is 2-5 cm. Its thickness there is 3-3 cm. (left bone). The breadth of the distal end is 7-4 cm. in the left and 7 cm. in the right bone. The distal end of both bones (text-fig. 22) is flattened and therefore broader than originally. A rounded ridge runs from the lower end of the crista radialis right to the distal edge of the condylus later- alis. A broad, low ridge passes from the condylus Text-fig. 22. Massospondylus Bvowni. Left humerus. medialis upwards on the Out!ine ?f dist^ ®nd. seen from below- Anterior , f. border above. Nat. size, shaft and disappears near the middle of the bone. Between these two ridges the anterior surface of the distal end is strongly concave. This concavity becomes narrower and shallower upwards and disappears in the middle of the shaft. The posterior surface of the distal end is concave along its middle, but it seems that this concavity is at least greatly exaggerated by pressure. The condyles are not produced backwards and they stand out slightly anteriorly. The articulation Annals of the Transvaal Museum 123 surface stands at right angles to the axis of the bone. The ulnar condyle is broad and not thick, the radial condyle is narrow and thick. Thickness of the ulnar condyle 2-5 cm., of the radial condyle 3-5 cm. Breadth of the ulnar condyle about 3 cm. and of the radial condyle 1-5 cm., all of the left humerus. Ulna and Radius. Both lower arms are present. Ulna and radius of the right arm have suffered badly from pressure. Those of the left arm have also suffered, but they are in better condition than those of the right. Ulna. The left ulna (PI. XVIII, figs. 2 and 4) has a length of 14 cm. The postero-medial edge of the bone has a sig- moidal shape, its upper end being convex and its lower end concave inwards and forwards; the antero-lateral edge is proximally concave outwards, in the middle straight or slightly convex and its lower end again concave towards the lateral side. The proximal end is slightly bent forwards and runs out an- teriorly into a sharp point. The posterior portion of the proximal end is much higher than the anterior portion, the difference in height being 1-7 cm. This portion is dome- shaped ; in front of it lies the concave articula- tion surface. The antero-medial border of the proximal surface (text-fig. 23) is sigmoidal. ft ulna.^OuUine of proximal end seen from above. Anterior end below and antero-medial border to the left. Nat. size. Its anterior end is convex; in the middle it is concave and its extreme posterior end is convex. The lateral border of the proximal surface is slightly concave, while the hinder border is convex backwards and upwards. The proximal surface has a triangular shape. The anterior angle is very sharp; the postero-medial angle, formed by the antero-medial border and the hinder border, is also acute, but broadly rounded, while the latero-posterior angle, formed by the posterior and lateral borders, is obtuse. The shaft is concave below the lateral border of the proximal surface. Below the posterior border the surface of the shaft is concave in one bone and convex in the other; it seems that the concavity is wholly due to pressure and that, therefore, the bone was originally convex in this region. In both cases the latero-posterior angle of ^ assos^on ' the proximal surface and the surface of the shaft for a short distance beneath it stand out from the bone backwards and outwards. The antero-medial border of the proximal surface has a length of 5*4 cm.> the lateral border a length of 4 cm. and the hinder border a length of 3-2 cm. The thickness of the bone, taken over the latero-posterior angle, is 2*4 cm. The antero-medial surface of the bone is concave from the antero-lateral border to the postero-medial border. Whether this was originally so in the middle of the shaft is not certain. The postero-medial edge of the whole bone is broadly rounded ; a large portion of its upper end is rough and served for the attachment of muscles. The upper part of the antero-medial surface is covered with short longitudinal grooves, which probably served for the dylus Browni. Left ulna. Outline of distal end seen from below. Medial end to the left, posterior border be- low. Nat. size. 124 Annals of the Transvaal Museum attachment of ligaments. At its narrowest place the shaft has a breadth of 2*3 cm. and a thickness of i-2 cm. The distal end (text-fig. 24) has a breadth of 37 cm. and a thickness of 1-9 cm. Its anterior end is turned to the lateral side and its posterior end to the medial side. Its long axis makes an angle of about 25 0 with the long diameter of the proximal end. The distal articulation surface has two facets. The postero-lateral part of the articulation surface passes somewhat upwards on to the shaft. The antero-medial edge of the distal end of the bone has two longitudinal ridges, separated by a groove. The postero-lateral ridge is high, the antero-medial one forms the slightly prominent border of the antero-medial surface. Radius. The radius (PI. XVIII, figs. 1 and 3) has a greatest length of 13-3 cm. The antero-lateral edge of the bone has a length of 12-3 cm. The proximal end of the radius is broad and thin (text- fig. 25). Its breadth is 3-6 cm. This was originally a few millimetres more, for the postero-medial corner of the proximal end is broken off. Its thickness is 17 cm. The proximal articulation surface is concave from the high posterior end of the antero- medial border to the slightly lower anterior end of the postero-lateral border. The breadth of the shaft at its narrowest place is i-8 cm. and its thickness 1*3 cm. The distal articulation surface has a length of 3-5 cm. (text-fig. 25) and a breadth of Text-fig. 25. Massospondylus Browni. i*9 cm. The fairly flat surface slopes from the postero-medial corner downwards to the antero-lateral end. Both the antero- lateral and the postero-medial edges are concave, but the antero-lateral one is convex at the distal end. This gives the impression that the distal end is bent slightly backwards and to the medial side. The distal end of the antero-lateral edge bears a sharp ridge, which runs from the articulation surface upwards for a short distance, and which leans forwards. The postero-medial edge of the distal end is fairly sharp. Left radius. Outline of proximal and distal end seen from above and below respectively. Anterior border above. Nat. size. Carpus. Two carpalia are preserved in both hands. Those of the right hand, how- ever, are fragmentary. A large flat bone was situated behind the first meta- carpale and is apparently the first carpale. A small bone was situated lateral to this one and was evidently displaced. It is most probably the second carpale. Both bones have been completely covered by cartilage. First carpale (PL XVII, fig. 5). The breadth of the bone is 4 cm. This may have been slightly greater, for the extreme upper medial corner is broken off. Its height is 2-5 cm., its medial thickness i-i cm. and its lateral thickness o-8 cm. The medial portion of the anterior surface is convex in its lower half, while its upper half is concave. The lateral portion of the anterior surface is also concave. The two concave surfaces are separated by a rounded ridge. The lateral concave surface occupies about one-third of the anterior surface. The posterior surface is convex. The upper edge of the bone is slightly da- maged. The medial edge is thick, and convex forwards, inwards and down- wards. The lower edge is broad and in its middle lies a pit, probably for the attachment of a ligament. The lateral edge is thin and projects towards the lateral side in its middle. The upper edge was thin. This bone can be fitted Annals of the Transvaal Museum 125 on to the proximal end of the first metacarpale in such a way that the upper edge of the bone lies at the same level as the upper edge of the proximal articulation surface of the first metacarpale. In this case the lower and the medial edges of the carpale project downwards beyond the lower edge of the articulation surface of the metacarpale. The ridge on the anterior surface of the carpale will then lie on the lateral proximal edge of the metacarpale, so that the whole concave lateral portion of the anterior surface of the carpale projects beyond the lateral side of the metacarpale. Second carpale (PI. XVII, fig. 6). The greatest dimension of the bone is 2*5 cm. When fitted in position, this dimension runs from above downwards and inwards. Its breadth is 13 cm. Its thickness below is 1 cm. and above it ends in a sharp edge. Its posterior surface is convex and its anterior surface concave. Its nearly square lower surface is flat and has a pit in the middle for the attachment of a ligament. The second metacarpale fits exactly with its convex posterior surface into the concave lateral portion of the anterior surface of the first metacarpale. Its upper end lies in this case near the lateral production of the lateral border of the first metacarpale, and its lower surface lies practically in the same plane as the lower proximal edge of the first metacarpale. The two ligament pits are then exactly opposite each other. When the second metacarpale is now fitted into place, it will be found that its proximal surface lies practically at the same level as that of the first metacarpale. The second carpale appears to be situated between the second metacarpale and the first carpale, while on its medial side it touches the first metacarpale and on its lateral side would probably touch the third carpale. Hand. Both hands are complete, but the left one is much better preserved than the right. The hand will be described as if it were stretched out with the palmar side downwards. Compare also PI. XVIII, fig. 5, PI. XIX, fig. 1 and text-figs. 26 and 27. Metacarpale I can easily be recognised from the others by its remarkable shape (text-figs. 26 and 27). The proximal articulation surface has a quad- rangular shape. The two lateral corners of this quadrangle are drawn out and far apart, while the two medial corners are rounded and close together. The shaft is very short. Two ridges run towards the distal end from the lateral corners of the proximal end. The ridge starting at the lower lateral corner is sharp and high and ends in the middle of the shaft. The superior lateral ridge is also high and sharp, but it runs right down to the distal end, where it branches around the lateral collateral pit. The lower medial corner of the proximal articulation surface is nearer to the distal end than other parts of the surface. The axis of the distal articulation surface is not parallel with that of the proximal surface. Its lateral end is higher and its medial end lower than in the proximal axis and moreover the lateral end projects much further distally than the medial end. The distal end is therefore twisted to the medial side. The two portions of the distal Text-fig. 26. Massospondylus Browni. Metacarpale I of the left hand. Outline view from above and slightly medial, showing the extent to which it was covered with cartilage. Nat. size. 9 126 Annals of the Transvaal Museum articulation surface are divided by a groove. The two collateral pits are deep. The articulation surface runs right round on the lower border of the medial collateral pit. The lateral portion is drawn out to a point above the lateral collateral pit and also below it. The lower surface is sharply concave immediately behind the groove dividing the portions of the articulation surface. Metacarpale II is the longest. Its proximal articulation surface is tri- angular (text-fig. 27). The whole surface is convex. Laterally it runs out into a narrow strip. This is the proximal face of a very high, thin ridge, which subsides on the shaft before reaching the distal end. A ridge runs from one of the corners of the proximal surface over the upper surface of the bone towards the distal end, and disappears before reaching the middle of the shaft. The lower surface of the proximal end has a broad, low ridge, which starts near the proximal surface and runs towards the middle of the shaft. There is a small concavity on both sides of this ridge. The inferior part of the distal end is much broader than the superior part, for the medial collateral pit opens upwards and inwards. Both collateral pits are deep. A division of the distal articulation surface is only indicated. The articulation surface passes much further on to the lower than on to the upper surface of the bone. The distal end is twisted somewhat to the lateral side. Text-fig. 27. Massospondylus Browni. Left hand. Outline in relative position of proximal ends of metacarpalia. Metacarpale V is deformed and could not be fitted into position. Nat. size. Metacarpale III. This is shorter and much more slender than II. Its proximal articulation surface is also triangular. The three ridges running from the corners of the proximal surface along the sides and upper surface of the shaft towards the distal end subside in the shaft before reaching its middle. The lateral portion of the lower surface of the proximal end is concave; its narrower medial portion is occupied by a broadly rounded ridge, which sub- sides in the middle of the shaft. The inferior part of the distal end is broader than the upper part, because both collateral pits open upwards. The articu- lation surface passes further on to the lower than on to the upper surface of the bone. The distal end is twisted somewhat to the lateral side. Metacarpale IV is more slender than III. Both its ends are relatively thicker. The proximal end has a triangular shape (text-fig. 27), but the upper angle does not lie in the middle of the upper surface as in the metacarpalia II and III, but on the extreme medial side. The upper angle is a thick rounded knob, which continues on to the shaft for a very short distance only. The lower medial angle is sharp and it sends a very short, sharp ridge forwards. The lateral angle is broadly rounded. The distal end is narrow and thick. The upper and lower breadth is practically the same. The shaft has a tri- Annals of the Transvaal Museum 127 angular section with a vertical medial side. The distal end is slightly twisted to the medial side. Metacarpale V is different in the two hands, but the left has suffered from pressure and the right has completely escaped this misfortune (text-fig. 27). The proximal end has a triangular shape of the same type as that of meta- carpale IV, only narrower and higher. The upper angle, which lies on the extreme medial side of the upper surface, is very broadly rounded. The lateral corner is also very broadly rounded. The lower medial corner is sharp and narrow; between this corner and the upper one lies a concave surface. The proximal articulation surface is greatly convex, except the lower portion, which is concave. The upper corner passes further forward on the shaft than any other portion of the articulation surface. The distal end is convex. Later- ally it runs out into a knob-like point; medially it is broadly rounded. A section of the shaft is more or less triangular, with a nearly vertical medial side and a convex lower one. The longest dimension of the distal articulation surface stands from above downwards and inwards. This metacarpale does not possess collateral pits. Phalanges. The first phalanx of the first digit is a remarkable bone. Its distal end is twisted nearly 45 0 on its proximal end. The proximal articulation surface has a trapezoidal shape. Its lateral side, however, is convex and its medial side is concave above and convex below. The lower medial corner of the surface is tongue-shaped and projects far to the medial side. The whole surface is deeply convex; it is divided into two unequal portions by a slight ridge; the lateral portion is the larger, the medial portion only consisting of a narrow strip along its concave border and of the tongue-shaped lower corner. The lower border is slightly concave below the end of the ridge over the articulation surface. The distal end is pulley-shaped. The groove between the two portions stands obliquely from above downwards and inwards. Its lateral side projects further forwards than its medial side, but the medial portion of the articulation surface passes further backwards on the lower surface than the lateral portion. The lateral collateral pit is deeper than the medial one. Excepting the claws, the phalanges of the second and third fingers have all the same general shape. The contour of the proximal surface of all these phalanges is trapezoidal. The proximal articulation surface of II, 1 and III, 1 is deeply hollow and not divided by a ridge. The upper posterior end of the phalanges does not project far backwards. The lateral side of their distal end projects further forwards than the medial side. The distal end of II, 1 is twisted to the lateral side, that of III, 1 slightly so. There is a slight con- cavity behind the anterior articulation surface on the upper surface of the bone. The lateral collateral pit is deeper than the medial one. As in the foot there is no pit behind the articulation surface on the upper surface of all the penultimate phalanges of the hand. The proximal articula- tion surface of both penultimate phalanges of the second and third fingers is divided into two portions by a ridge. The lateral portion is broader than the medial one. The penultimate phalanx of the second finger differs from that of the third, except in size, through the upper portion of the medial border of the proximal surface being concave in the former and straight in the latter. A further difference is, that in the penultimate phalanx of the second finger the lateral side of the distal articulation surface passes slightly further down- wards than the medial side, whereas in the penultimate phalanx of the third finger this is just the reverse. Both phalanges have a broad and thick upper posterior process. The distal articulation surface of both phalanges passes 9—2 128 Annals of the Transvaal Museum much further backwards on the lower than on the upper surface. Their lateral collateral pit is deeper than their medial one. The second phalanx of the third finger has a divided proximal articulation surface. There is no pit behind the distal articulation surface on the upper surface of the bone, and contrary to the condition in the penultimate phalanges of the second and third fingers, where one side of the distal end does practically not project beyond the other, the lateral side of the distal end of this phalanx projects beyond its medial side. The lateral collateral pit is deeper than the medial one. The lateral portion of the distal articulation surface of II, i, III, i and III, 2 is broader than the medial portion. The fourth and fifth fingers are remarkable. The fourth has only three phalanges and the fifth only two. The phalanges of both fingers can be re- cognised by their plainness. None of them has a superior-posterior process. The first phalanx of the fourth finger has an evenly concave proximal articu- lation surface. The distal articulation surface is evenly convex; there is no pit behind it on the upper surface. The second phalanx has an evenly concave proximal and an evenly convex distal articulation surface. There is no pit on the upper surface. The collateral pits are represented by very slight con- cavities, bordered below by a knob-like process. The endphalanx of this finger is a remarkable, small, three-edged body. Two of the edges border the very slightly concave articulation surface. The third edge forms the sides and the anterior end of the bone. The first phalanx of the fifth finger has an oval-shaped evenly concave, proximal articulation surface. It stands some- what obliquely on the axis of the bone, the superior border lying further forwards. The distal articulation surface is evenly convex and its lower lateral side sends a small process backwards below the collateral pit. The endphalanx of the fifth finger has the same general shape as the endphalanx of the fifth digit of the foot. Its proximal articulation surface is oval-shaped and very slightly concave. It stands obliquely on the axis of the bone for its superior edge lies much further forwards than its inferior one. The inner edge of the bone is straight and the outer-anterior one evenly convex from the inner-anterior end to the outer-posterior end. The anterior portion and part of the outer portion of this edge has been covered with cartilage. The clawphalanx of the first finger is high and strongly curved (PI. XVII, fig. 4). The groove for the claw lies slightly deeper on the medial than on the lateral side. There is a broad ridge on each side below this groove. The highest part of this ridge on the medial side lies below its middle line, while the highest part of the lateral ridge follows the middle line of the ridge. The proximal ends of both ridges terminate some distance in front of the articula- tion surface by a low, sharp, curved ridge. The medial ridge passes further backwards than the lateral one. The distal end of the medial ridge lies slightly higher than that of the lateral one. There is a broad thick "boss on both sides in front of the middle of the articulation surface. The boss on the lateral side is thicker and is situated higher up the side of the bone than the medial one. The bosses are separated from the just mentioned ridges on their respective sides by a broad groove. These grooves pass downwards and backwards above the boss for the flexor tendon, where they become deep and narrow, and con- tinuing come very near to each other behind it. Upwards they are continuous, with the grooves for the claw. The tuberositas for the flexor tendon is divided into two portions by a longitudinal groove, which lies to the lateral side of the middle line on the lower surface. The lateral portion of the tuberositas, which lies higher on the lateral side than the inner portion on the medial one, is high and narrow, while the inner portion is low and broad. As in all these Annals of the Transvaal Museum 129 clawphalanges, the articulation surface is divided by a vertical ridge into two parts, of which in this case the medial one is slightly broader in its lower end than the lateral one. Through the position of the distal articulation surface of the preceding phalanx, the clawphalanx must lie obliquely from above downwards and inwards. The clawphalanges of the second and third fingers are much less curved. The lateral groove for the claw is slightly higher than the medial one, although the difference in height is scarcely noticeable in the third clawphalanx. Measurements of metacarpalia and phalanges of the left hand in centi- metres (for the fifth metacarpale those of the right hand have been sub- stituted) : Length Proximal Distal Breadth Thickness Breadth Thickness I 4-2 3-4 2-4 inf. 3‘4 sup. 3-3 i*7 I, I 3*5 3-2 2-7 2*2 1-2 2-1 I, 2 6-3 i-6 3-4 — — — II 5-6 3*2 i-6 2-6 i*7 i-8 II, I 2-8 2-5 2 2-1 1’4 i-6 II, 2 2-7 i-9 i*9 1-7 1 i-6 IT 3 4-6 1‘3 2 — — — III 5-i 2-6 1*2 2 i*3 1-2 III, I 2-3 i*9 i‘4 i-8 1-2 I-I III, 2 i*9 i*6 i*5 i-6 I-I I-I III. 3 i-8 i*4 i-3 i-3 o-8 I III, 4 3-i i-4 1 — — — IV 4‘3 i-8 i-6 i*5 i*3 I-I IV, 1 i-8 i*4 1 1-2 1 0-9 IV, 2 1*2 o-9 0-9 I o-6 o-6 * IV, 3 o-6 o-7 o-5 — — — V 2-8 1-7 i-6 i'5 1 V, i i-6 i-3 o-8 1 o-6 V, 2 1 o-7 o-5 - - 0-3 A few millimetres are missing from the extreme end of the clawphalanx of the first finger. The length of its lower surface has been taken for the length of the ultimate phalanx of the fourth finger. Ileum. Both ilea are preserved, but they have suffered greatly from pressure. The ileum is a broad plate with a short spina anterior and a long spina pos- terior (PI. XIX, fig. 2) . The acetabulum cuts deep into the bone and is bordered in front by a long processus praeacetabularis and behind by a short processus postacetabularis. The upper portion of the ileum is thin, the lower portion thicker. Both spina anterior and posterior are directed outwards and the outer surface of the bone is therefore concave. The length of the bone from the spina iliaca anterior to the spina posterior is 19-9 cm. in the left and nearly 21 cm. in the right ileum. The upper border, which is damaged, is convex. The outer surface of both spinae and of the upper border as far as preserved is coarse, and has apparently served for the attachment of muscles. This coarse surface is especially large on the spina posterior. The hinder end of the spina posterior is truncated. Nothing can be said of the medial surface of the bone, for in both cases it is covered by matrix and other bone material, which could not be removed. The height of the ileum, from the processus I3° Annals of the Transvaal Museum postacetabularis to the upper border, is 13 cm. in the left and 13-5 cm. in the right ileum. The acetabular notch has a depth of 5 cm. and a breadth of 10 cm. in the left ileum. The crista supra-acetabularis projects to the lateral side as a sharp, thin ridge from the proximal two-thirds of the processus praeacetabu- laris. Probably it continued right down to the distal end of the processus praeacetabularis. This ridge becomes thicker and lower on the iliac plate, where it dies out before reaching the processus postacetabularis. The ace- tabular surface has a breadth of 3 cm. in the left ileum and of 4 cm. in the right one. It is concave in all directions. The inner border of the acetabular surface is a sharp ridge . The processus praeacetabularis reaches further forwards than the spina iliaca anterior. There is a deep notch between the two, which is much wider in the right than in the left ileum, probably a result of pressure. The length of the processus praeacetabularis is 9-5 cm. There are two sacral ribs (?) on the medial side of the right ileum. The posterior one is situated on the hinder border of the bone, and both have been pressed into the bony matter of the ileum. They are just where one expects the crista medialis. The acetabulum of the left ileum gives one the impression that in the natural position of the bone the end of the spina iliaca anterior was situated nearly perpendicularly above the head of the femur. Pubis. Both pubes are present, but in a very bad condition. The distal portion of the right pubis is missing. The processus subacetabularis and the down- ward bent portion at the upper inner corner of the pubic plate of both pubes are missing (PI. XIX, fig. 3) . When found the pubes were lying parallel and near to each other, but there was no bony connection between the pubic plates. However, it seems not improbable that the pubic plates were originally coalesced. The medial borders of both plates are fractured over their whole length. Although very much flattened, the right pubis still shows that the neck has an anterior and a medial surface, which round off broadly into each other. The lateral posterior surface of the neck is convex. The inner border of the neck is sharp. The shape of the neck is, therefore, essentially the same as that of the pubis of Dromicosaurus gracilis . The length of the right pubis as presefved is 28 cm. The breadth of the pubic plate in the middle, as pre- served, is more than 4 cm. Ischium. The proximal portions of both ischia are present. They are completely flattened out and so crushed generally, that it will suffice to figure them (PI. XIX, fig. 4). It seems, however, that the articulatio iliaca made a much larger angle with the general long axis of the bone than in Dromicosaurus gracilis. Femur. Both femora are present, but they are badly crushed. The left femur (PI. XX, figs. 1 and 2) was broken in several places, but the pieces were fitted together and joined with plaster of Paris by myself; these fractures have had no influence on the length of the bone. Both bones are curved sigmoidally, concave above and convex below towards the front. Most of the curvature, however, has been crushed out of the right femur. The length of both femora is 35 cm. The breadth of the proximal end, measured from the tip of the caput femoris to the lateral side, is 10 cm. in the left and 9-5 cm. in the right femur. These measurements are of course very much exaggerated through the crushing of the bone. The original breadth can easily have been 1 cm. less. The caput femoris is directed inwards; as preserved its height is about 4 cm. and its Annals of the Transvaal Museum 131 breadth about 2 cm. ; originally the breadth may have been greater than the height. The proximal surface is convex and forms an angle with the lateral side. The trochanter minor lies on the hinder surface at the border of the proximal surface and nearer to the medial end of the caput femoris than to the lateral side of the bone. The upper end of the trochanter major is broken off in both bones. It was situated at a distance of 6*2 cm. from the proximal end in the left femur; this distance is 5-5 cm. in the right bone. There is a broad, deep groove between the lateral side of the trochanter and the general surface of the bone. The height of the trochanter, as preserved in the left femur, is 9 cm. The lower end of the trochanter is not visible. Between its proximal end and the trochanter major the femur was broader and thinner than further downwards. Both bones are so badly crushed that nothing can be seen of the ridge, which in other femora runs from the trochanter major towards the condylus medialis. The lateral side of the proximal end is slightly convex and the medial side concave. Therefore the proximal end of the femur is slightly bent inwards. The trochanter quartus begins at a distance of 9 cm. from the proximal end in the right femur. This distance is 10 cm. in the left one. It terminates at a distance of 15-5 cm. in the right and 16-5 cm. in the left femur from the proximal end of the bone. This lower end lies at a distance of 18-5 cm. in the left and 19*5 cm. in the right bone from the distal end. The height of the trochanter is 2 cm. Its medial side is steep and its lateral side less so. The proximal end of the trochanter forms a very pronounced ridge ; a second ridge, medial to this first one, is only indicated by a slightly greater convexity of the lateral side of the trochanter. The trochanter is situated nearer to the medial than to the lateral side of the bone. Its lateral side is convex and its medial side concave. The fossa intercondyloidea begins at about 11 cm. above the distal end. Its upper end lies nearer to the medial than to the lateral side of the bone, and its lower end is slightly further away from the medial side. The condylus medialis has a breadth of 2-8 cm. in the left femur and a breadth of 2*4 cm. in the right one. In the right femur the condyles have been pressed very much to the medial side and flattened; in the left one they have been slightly pressed to the lateral side. The height of the condylus medialis is 3-8 cm. It projects about 3 cm. from the bottom of the fossa intercondyloidea. The condylus lateralis has a breadth of 2-4 cm. in the right femur. Its height is about 3 -5 cm. and it projects about 2-5 cm. from the bottom of the fossa intercondyloidea. The breadth of the distal end of the bone is about 9 cm. in both femora. The distal articulation surface stands obliquely on the axis of the bone, that is to say, if this, surface is placed horizontally, then the axis of the bone is directed from below upwards and backwards. The anterior surface of the lower end is slightly concave. The lateral surface is hollow and the condylus lateralis stands out slightly to the lateral side. Tibia. Both tibiae are preserved, but in a very bad condition. The proximal end of the right tibia (PI. XX, fig. 3) is distorted and that of the left flattened to less than finger thickness. As preserved, the length of the bone is 30-5 cm. The head of the left tibia has now attained a length of about 12 cm. and a thickness of 1-3 cm. at its anterior end and of 3 cm. at its posterior end. The head of the right tibia has a length of 8-5 cm. measured from the hinder end of the condylus lateralis (!) to the anterior portion of the medial border. Its breadth is 5 cm. I give these measurements, which are of no value to our knowledge of the animal they belong to, for reasons which will be discussed 132 Annals of the Transvaal Museum hereafter. The tuberositas tibiae of the right tibia is still visible. The lateral condylus of the right tibia is flattened and now forms a kind of lateral posterior process. The concave portion in the lateral border of the proximal articulation surface is slightly deepened, but the notch in the hinder border has been greatly exaggerated through pressure. The articulation surface slopes from behind upwards and forwards. The shaft becomes narrower from the head downwards. The distal end is transversely broad; the bones are so badly crushed that it is useless to say more about them. As far as can be made out, they have the same general shape as other Thevopod tibiae. Fibula. Of the left fibula only the proximal end is present. The right fibula is complete, but has suffered much from pressure. The original curvature of the bone is completely flattened out. The fibula has the same general shape as the fibula of Dromicosaurus gracilis. Its length is 30 cm. The high ridge on the shaft of the fibula of Dromicosaurus gracilis is still visible in this specimen as a faint low ridge. The antero-medial corner of the distal end is broken off. The fibula is figured in fig. 4 of PI. XX. Tarsalia. Fragments of the tarsalia of the left and the right side are present, but they are too badly preserved to study in detail. Foot. The right foot is complete (PL XXI, fig. 1 and PL XXII), and of the left foot the first two toes are present. All the bones, and especially the metatar- salia, have suffered from pressure. Metatarsale I has a length of 9-3 Cm. in the right and 9-6 cm. in the left foot. The bone of the right foot is more flattened than that of the left. As preserved the proximal breadth of the right bone is 4 cm. and that of the left 3-6 cm. The thickness of this end is 1*2 cm. and 1*3 cm. respectively. Posteriorly the proximal end of the bone has a broad rough edge, which passes further downwards into the smooth posterior surface of the shaft. There is a slight angle between the rough and the smooth edge. Proximally the lateral surface bends towards the medial side near the posterior border. Nothing is visible of a ridge on the medial side of the proximal end, but this may be due to pressure. The shaft narrows down to a breadth of 2 cm. and 2*4 cm. and a thickness of 1-5 cm. and i-i cm. in the left and right bones respectively. The axis of the distal articulation surface stands obliquely on the axis of the bone ; the medial posterior end of the surface is much higher than the lateral anterior end. Further the axis is directed from the front backwards and somewhat to the medial side. The antero-medial part of the articulation surface is undivided. Below and behind it consists clearly of two portions, a large, globose, antero-lateral one and a narrow, transversely elongated, medial portion, which sends a long narrow process upwards, backwards and to the lateral side. There is a deep concavity between the globose portion of the articulation surface and its hook-like process on the postero-lateral surface of the bone. On the under surface the two portions of the articulation surface are divided by a very shallow, broad groove. Metatarsale II has a length of 14 cm. The proximal articulation surface has the shape of a quadrangle. All four sides of this quadrangle are concave. The deformation makes it impossible to say which side of the bone is more concave than the other. The lateral border of the proximal surface has a length Annals of the Transvaal Museum 133 of 3-6 cm., the medial border a length of 3-9 cm., the posterior border a length of 3-1 cm. to 3-3 cm. and the anterior border a length of 2-6 cm. Ridges run from the corners of the proximal surface downwards on to the shaft. The upper end of the antero-lateral ridge is very thin and sharp; the antero- medial ridge is rounded ; the two posterior ridges stand out prominently but are deformed. All these ridges submerge into the rounded edges of the shaft. There is a small tuberositas on the lateral border of the anterior surface near the middle of the shaft. At its narrowest part the shaft has a breadth of 2-5 cm. and 2-3 cm. and a thickness of 1-5 cm. and 1-7 cm. in the right and left bones respectively. The distal articulation surface does not stand at right angles to the shaft, its medial end being higher than its lateral end. As pre- served, the breadth of the distal end of the right bone is 3-8 cm., its lateral thickness r6 cm. and its thickness at the medial side 2-2 cm. The medial thickness of the distal end of the left bone is 2-5 cm. The articulation surface passes higher up on the medial side of the anterior surface of the bone than on the lateral side. Posteriorly the medial side of the distal end has a back- wardly directed process, the lower surface of which is a backwardly directed process of the articulation surface. The middle of the posterior surface of the distal end is slightly concave. The greatly elongated lateral collateral pit is much deeper than the medial one. The anterior surface of the distal end is smooth. Metatarsale III has a length of 15-5 cm. The proximal articulation surface has a triangular shape. The bone is very much flattened, but as preserved, the hinder angle lies towards the medial side. The two anterior angles lie close together. As the medial anterior ridge is broken off the lengths of the medial and anterior borders of the proximal surface cannot be given, but as preserved they are each approximately 2-5 cm. The lateral border has a length of 4-7 cm. The two medial ridges converge downwards and subside into the medial side of the shaft. At this point the medial side of the shaft is slightly convex. The lateral side of the shaft shows a small tuberositas a short distance above this medial convexity. The lateral ridge disappears earlier into the lateral side of the shaft. At its narrowest part the shaft has a breadth of 2-3 cm. and a thickness of 1-3 Cm. The axis of the distal articulation surface does not stand at right angles to the shaft, its medial end being higher than its lateral end. The breadth of the distal end is 3-6 cm., its lateral thickness 1-7 cm. and its medial thickness 1-9 cm. The middle of the posterior surface of the distal end is concave. Posteriorly the medial side of the distal end has a backwardly directed process, the lower surface of which forms a backward prolongation of the articulation surface. The anterior surface of the distal end shows a peculiar feature. The centre above the articulation surface is convex. Around and above this convexity there is a half-moon shaped con- cavity. Above this is the flat anterior surface of the shaft. Both collateral pits are deep, but the lateral one is deeper. Metatarsale IV has a length of 14 cm. As preserved the proximal portion is very thin; its anterior surface is convex and its posterior surface is concave. Its greatest thickness here is i-i cm. and its breadth 6 cm. The bone has suffered greatly from pressure and is quite out of shape. A ridge runs down its anterior surface and starting near the middle of the upper end reaches the medial side above the middle of the shaft. At this point it forms an elongated knob on the side of the shaft. At its narrowest part the shaft has a breadth of 2-4 cm. and a thickness of 1 cm. The lateral end of the distal articulation surface is situated much higher than the medial end. Anteriorly the breadth of the articulation surface is 2-5 cm., posteriorly it is 3*1 cm. The reason of 134 Annals of the Transvaal Museum this difference is, that the hinder border of the lateral collateral pit stands out very far laterally. There is no medial collateral pit, the medial surface of the distal end being evenly concave. Posteriorly the medial side of the distal end forms a sharp ridge, the lower surface of which joins up with the articulation surface. The posterior surface of the distal end is concave along its middle. Metatarsale V. A small piece of the distal end of metatarsale V is broken off. As preserved the whole length of the bone is 7 cm. The whole length may have been 7-5 cm. Proximally the bone is very broad and thin. Its breadth here is 4-3 cm. Laterally the thickness of the proximal end is 1-5 cm., but towards the medial side it rapidly thins down to 0*7 cm. Its medial edge is sharp. The medial end of the upper border of this thin portion is bent slightly backwards; the medial border runs downwards, forwards and outwards. The anterior surface is convex, the posterior concave. The hinder lateral ridge is sharp and runs downwards dnd to the medial side. At the lower end of this ridge the bone has a breadth of 1*5 cln. and a thickness of 1-3 cm. A broad low ridge starts at this spot on the lateral surface and runs downwards and forwards. All the phalanges have been preserved. In the following table their measurements are given in centimetres : Length Proximal Distal Breadth Thickness Breadth Thickness I, I 5 3-4 2-3 ant. i-5 post. 2-8 2-1 I, 2 7-8 2-2 3'4 — . — — II, I 6 3’3 2-4 2-6 3-4 i-9 II, 2 4 2-9 2-3 i-7 2*5 i*9 II, 3 ±6-8 i-8 2-8 — - — - III, I 6 3*7 2-1 2-5 3'3 i*7 III, 2 4’3 3 2 2 2-8 i-6 HI. 3 3-5 2*6 i*8 i*5 2-3 i-6 III, 4 1 — 2*1 2 — : — — IV, 1 4-6 3*2 i-8 2 3 i-5 IV, 2 3'5 2-8 i*8 i-8 2-7 i*4 IV, 3 3 2-6 i'7 i-6 2-4 i*4 IV, 4 2-7 2-2 i-6 1-2 2-1 i-3 IV, 5 5*i i*7 i-8 — — — v, i i-8 i-5 o*8 — — o-5 Remarks. The anterior breadth of the distal end has been measured across the upper anterior boundary of the articulation surface. The first phalanges can be recognised by the fact that their proximal articulation surfaces are evenly concave and that the hinder border of this surface is straight, at all events not projecting upwards in the centre. All the penultimate, phalanges have in common that the distal ends of their anterior surfaces are evenly concave and not pitted. These two characters, therefore, separate I, 1 from all other phalanges. Moreover the distal end of I, 1 is twisted slightly* to the lateral side. The posterior border of the proximal articulation surface is higher than the anterior one. The distal articulation surface is unsymmetrical and the articulation ridges of this end run much further upwards on the posterior than on the anterior surface. The proximal articulation surface of the second, third and fourth pha- langes and claws is divided into two parts by a ridge and corresponds with Annals of the Transvaal Museum 135 the pulley-shaped articulation surface of the preceding phalanx. The middle of both its anterior and its posterior border projects upwards, however an- teriorly more strongly than posteriorly. This anterior upper process is strong- est in II, 2. The first claw is more bent than the others. All the claw- phalanges are unsymmetrical. The medial portion of the articulation surface of clawphalanges I, II and III is smaller than the lateral portion. In claw- phalanx IV I could see no difference in size. Clawphalanges I, II and III hang over towards the lateral side. Clawphalanx IV hangs slightly over towards the medial side. In clawphalanges I and II the edge between the lateral surface and the lower surface is rounded, while the edge between the medial surface and the lower surface is a sharp ridge. In clawphalanx III both edges are sharp, although the medial edge is sharper. In clawphalanx IV the edges are narrow rounded ridges, the one exactly like the other. In claw- phalanges I — III the groove for the claw lies deeper on the medial than on the lateral side. In clawphalanx IV the medial groove may be very slightly higher than the lateral one. The phalanx of the fifth toe has a remarkable shape. It is a flat bone, the shape of which is generally triangular. What is probably the proximal end has a breadth of i-6 cm. and a thickness of 0-9 cm. What is probably its medial side has a length of i*8 cm. and is concave. If this is the right position of the bone, which I do not doubt, for the endphalanx of the fifth finger has this position and has the same general shape as the present bone, the anterior and lateral sides form an evenly convex border. The antero-lateral surface is smooth and gives the impression of having been covered with cartilage. Its posterior border goes slightly further down than its anterior border. The anterior and posterior surfaces of the bone are concave. Discussion. While comparing the bones of the lower arm with those of the Plateo- sauvidae as described by v. Huene (5), it was found that there were certain differences which required an explanation. There is one point which can be decided without difficulty and that is, which of the three angles of the tri- angular head of the ulna is the anterior one. The articulation surface is of course anterior to the high olecranon-like process and the angle below it must be the anterior angle. This angle lies between the two long sides of the triangle. If the ulna of both arms of our specimen is placed with the anterior angle forwards, it will be found that the longest side of the triangular head lies on the medial side. According to description with text-figure of ulna and radius of Plateosauvus Reinigeri v. Huene it seems to be different at least in this species. Here the longest side of the triangular head lies on the lateral (antero- lateral) side of the bone, and the angle between the two long sides of the triangle lies on the lateral side. The figured right ulna of Plateosauvus Reinigeri would have to turn 90° on its axis to place the narrow angle of its upper end in front of the remainder of the head. If the narrow angle of the ulnar head of Plateosauvus Reinigeri were pointing forwards, its contour would be placed in the same position as the contour of the head of our right ulna and one would therefore conclude that the described ulna of Plateosauvus Reinigeri is also a right one, which it actually is. Let us now compare the ulna of our form with the description and text- figures of the ulna of Plateosauvus Quenstedti v. Huene. In the text (/. c. p. 36) it is stated that the anterior angle lies between a lateral side (the longest) of 9 cm. length and a medial side of 7 cm. length. The posterior side has a length of 6 cm. An upper view is given of the left ulna in the text-figure. Its 136 Annals of the Transvaal Museum anterior angle is therein directed downwards, the upper horizontal border is the posterior side, while the long lateral side of the head lies on the right-hand side of the figure. If an upper view of our left ulna is placed alongside of this text-figure, it will be seen that while the anterior angle is directed downwards and the hinder border lies away from the reader, the longest border is situated on the left-hand side of the figure. However, if an upper view of the right ulna of our specimen be placed alongside of the upper view of the left ulna of Plateosaurus Quenstedti , it will be seen that the two figures are identical. Moreover, in an adjoining text-figure the lateral side is shown of the left ulna. This lateral side is identical with the medial side of our right ulna. These considerations would lead to the conclusion that, what is supposed to be the left ulna of Plateosaurus Quenstedti is really the right one. With the ulna of Plateosaurus erlenbergiensis v. Huene it is the same thing. What is described as the head of the left ulna corresponds with the head of our right one. What is called the right ulna of Pachysaurus ajax v. Huene on p. 143, l.c. would be the left one according to our specimen. From the above may be concluded that the shape of the head of the right ulna of Plateosaurus Reinigeri is not identical with the contour of any of the mentioned right ulnar heads; that it would be identical with the contour of the right ulnar head of our form if it were turned 90° on its axis; that the contour of the left ulnar heads of Plateosaurus Quenstedti, Plateosaurus erlen- bergiensis and Pachysaurus ajax is identical with that of the right side of our form. On p. 59 l. c. v. Huene gives a figure and a description of the left radius of Plateosaurus erlenbergiensis ; the figure presents a medial view. The anterior portion of the medial border of the proximal end and the posterior portion of its lateral border are shown to be high. A medial view of the left radius of our specimen shows the posterior part of the medial border and the anterior part of the lateral border to be high. The contour of a reflected image of the lateral side of our left radius is, however, identical with the contour of the medial side of the left radius of Plateosaurus erlenbergiensis ; or, in other words, the contour figure of the medial side of the left radius of Plateosaurus erlen- bergiensis is identical with the contour of a lateral view of our right radius. These considerations would therefore lead to the conclusion, that what is styled left radius of Plateosaurus erlenbergiensis is really the right one. Dr Broom gave a figure of the left ulna of M assospondylus Harriesi (7, PI. XVI, fig. 15) and it clearly shows that the long side of the triangular head lies medial, while the angle between the two short sides lies on the lateral side. This is therefore in full agreement with the ulnae in our specimen. I would like to point out that our specimen was taken out of the matrix by myself and that no interchanging of bones could have taken place. The left arm was out of the matrix and every bone labelled and packed away long before the right arm was discovered. Moreover, weathering agencies have had a different effect on the two arms, all the bones of the left arm now having a reddish colour, while all those of the right arm are white. Besides, the bones of the left arm have only suffered slightly from pressure, while nearly all the bones of the right arm are badly crushed. Therefore, even apart from my statement, there is every reason to believe that the bones did not get mixed up. I have given the measurements of the heads of the tibiae, because I would like to prove the folly of describing fragments in such a crushed and flattened condition under new generic and specific names. If the heads of the above described tibiae had been found separate one could never have proven them to belong to the same nor to different species. However, for example Euskele- Annals of the Transvaal Museum 137 saurus capensis Lydekker sp. is based on a fragment of a tibia, which is crushed and flattened out of recognition. Nobody will ever be able to prove that this species differs from Euskelesaurus Browni. The only result of such work can be, that science is for ever burdened by meaningless names and synonyms. With regard to the affinities of this specimen, it will be clear from the different ilea that it does not belong to Gryponyx. Comparison with Massospondylus carinatus shows immediately that its ileum has practically the same shape as that of our form. Length and height of the bone in our form are 20 cm. and 13 cm. and in Massospondylus carinatus 22 cm. and about 14 cm. The relations in the two bones are therefore also practically the same. Our individual may, however, have been slightly smaller. Pubis and ischium of our specimen are too much damaged to allow of comparison with those bones of Massospondylus carinatus. The femur portion below the trochanter quartus in Massospondylus carinatus measures about 22 cm. and the distance of its trochanter major below the proximal end is 9 cm. In the femur of our specimen these distances are 18-5 cm. and 6*2 cm. Relatively, therefore, the trochanter major lies considerably higher in our form than in Massospondylus carinatus. The tibia of the present specimen is too much crushed to be used in a comparison. The humerus of Massospondylus carinatus is only known from fragments, and radius and ulna are altogether unknown. Metacarpale I of Massospondylus carinatus is rela- tively slightly broader than that of our form. The first phalanx of the second finger is proportionately longer than in our form and the fifth metacarpale is proportionately broader. The first phalanx of the second toe of Massospon- dylus carinatus has the same length as that of our form but is much broader. Comparison with Massospondylus Harriesi (7, p. 299). The humerus of this type seems to be slightly longer than that of our specimen. At any rate, the distance from the lower end of the pectoral ridge to the furthest part of the distal end is 12-5 cm. in Massospondylus Harriesi, whereas it is 11*5 cm. in our form (11 cm. up to the distal end of the condylus lateralis). Radius and ulna of Massospondylus Harriesi, however, are both slightly shorter than in our form. The first metacarpale of our form is absolutely longer and narrower than that of Massospondylus Harriesi: In Massospondylus Harriesi the first phalanx of the first digit is longer than its metacarpale, in our form this is the reverse. The first claw of Massospondylus Harriesi is longer than that of our form. There are numerous small differences in the other phalanges of the hand. The distal portion of the femur of Massospondylus Harriesi, measured from the lower end of the trochanter quartus, is 15-5 cm. In our form this portion measures 18-5 cm. The metatarsalia of Massospondylus Harriesi are all shorter than those of the present specimen, but where metatarsale I is more than 6 mm. shorter, metatarsale II is only 2 mm. and metatarsale III only 1 mm. shorter. The relations are therefore different. All the phalanges of the foot of our form are longer than those of Massospondylus Harriesi, but •also relatively more slender. The relations of the humerus of Aetonyx palustris are practically the same as in our form, but the radius is relatively smaller. Relative to the first metacarpale, the first claw of Aetonyx palustris is much longer than in our form (the distal width of the first metacarpale as given by Broom (7, p. 305, PI. XV, fig. 12)). In Aetonyx palustris the first metacarpale is practically as broad as long; in our form the breadth is much less than the length. In Aetonyx palustris the second phalanx of the second finger is longer than the first and the third phalanx of the third finger is longer than the second. Annals of the Transvaal Museum 138 In our form this is the reverse. Similar differences can be found in the foot. The lengths of the metatarsalia of Thecodontosaurus shirtopodus show rela- tions to each other which differ from those of our form. Only the femora are known of Thecodontosaurus Browni. If the femur of Thecodontosaurus Browni and that of the present specimen are given the same length as that of Dromicosaurus, and the lower end of the femur of Masso- spondylus*carinatus be given the same length as the lower end of the Dromico- saurus femur, then the other measurements become as tabulated below (in centimetres) : Dromicosaurus gracilis Thecodontosaurus Browni Present specimen Massospondylus carinatus These measurements show that there is a fairly big difference between the femora of Dromicosaurus gracilis and of Thecodontosaurus Browni, and also between those of our present specimen and of Massospondylus carinatus. The difference, however, between the femur of Thecodontosaurus Browni and that of the present form is very slight. The difference is so slight that, together with the consideration that our specimen is badly preserved, I do not feel myself justified in ascribing them to different species. The genus to which they belong cannot be Dromicosaurus. Except the difference in the femora, there is a great difference in the relation of femur and tibia. The tibia is relatively much longer, with regard to the femur, in the present form than in Dromicosaurus . As preserved, the articulatio iliaca of the ischium of the present form makes a much larger angle with the general long axis of the bone than that of Dromicosaurus. It is difficult to conceive that this could be the result of pressure alone. The proximal hollow surface between the articulatio iliaca and the articulatio ischio-pubica seems to be relatively larger in Dromicosaurus . Could the required genus be Massospon- dylus? It is difficult to say. The ileum of Massospondylus carinatus is practi- cally the same as in the present form, but the trochanter major lies relatively so much lower in Massospondylus carinatus. However, this is the only prin- cipal difference which I could find. Considering the bad condition of the present specimen I think it will be better to place it in the genus Massospon- dylus and keep it there till it can be definitely proven to belong somewhere else. Some time ago our collector at Harrismith found some Dinosaurian remains on the commonage. They consist of a vertebra and some phalangeal bones, and although they could not be identified as to genus or species, they are here described because they are so well preserved and because the vertebra shows a remarkable vertical ridge, running upwards from the articulation surface for the capitulum costae. There is no indication whatever that these bones belong to one individual. Length Proximal end to . trochanter major Proximal end to lower end trochanter IV Lower end trochanter IV to distal end 49-5 7'5 21-5 28 49-5 8-25 227 26-8 49-5 877 23-34 26-16 — 11 -5 — 28 Annals of the Transvaal Museum 139 Vertebra. The vertebra is the first dorsal (PI. XXIII) . The length of its centrum near the neural suture is 7-2 cm. The distance between the lower ends of the articu- lation surfaces is 7*7 cm. The height of the anterior articulation surface 6-7 cm. and its breadth is about 5 cm. The height of the posterior articulation surface is 6-2 cm. and its breadth 5-4 cm. The centrum possesses a tremendously narrow keel, the lower border of which is nearly straight. The articulation surface for the capitulum costae lies slightly in front of the middle of the centrum and borders on the centro-neural suture. It also lies nearly straight below the large concavity under the processus transversus. It is a deeply concave surface, which is bordered by a high wall, and which is open behind. Its hinder margin, however, lies higher than the nearest surface of the centrum and is therefore a ridge. This ridge continues upwards and through its sig- moidal shape its upper end lies straight above the middle of the parapophysis. Here it is prominent, but its height diminishes as it continues until it dis- appears well between the two downward ridges from the processus transversus and nearer to the anterior than to the posterior one. There is a small knob immediately below the parapophysis. Both transverse processes have been broken off, the left one near its origin, and the right one in such a way that the hinder border of its distal end is still preserved. The processus is directed slightly backwards and what is left of its upper surface slopes outwards and slightly upwards. The processus is supported below by two ridges, which diverge downwards under an acute angle. The outer edge of the posterior ridge is thicker and stands further outwards than that of the anterior ridge. The distal end of the processus may have been triangular and is much thicker than its proximal end. The lower angle of this triangle is prolonged into a small, downwardly directed knob. The spaces between the processus trans- versus and the zygapophyses are roofed in by a thin sheet of bone. The space between the supporting ridges of the processus and these bony roofs is occupied by deep pits, the hinder one of which is deeper than the anterior one. The whole upper surface of these bony roofs and of the processus transversus, which is flat as far as preserved, slopes forwards and downwards. There are also supporting ridges below the zygapophyses. All the supporting ridges of one side form together the letter W. The praezygapophyses are broken off. The postzygapophyses project beyond the articulation surface of the centrum. Their articulation surfaces converge downwards under an acute angle. There is a deep groove between the postzygapophyses which runs forwards and cuts into the hinder edge of the processus spinosus. The processus spinosus is very narrow and thick; its upper end is damaged. Both articulation surfaces of the centrum are concave, however, the hinder one more so than the anterior one. The neural canal has an anterior height of 2-1 cm. and breadth of 2-6 cm. and a posterior height of 2-4 cm. and breadth of 1-9 cm. In the middle its height is about 4*5 cm. and its breadth about 1-9 cm. Foot. There is a proximal end of metatarsale III. The proximal surface is slightly concave. The anterior border is broken off. The medial border is fairly straight and slightly convex in its hinder portion. Length of medial border 5-2 cm., of anterior border 3-2 cm. and of lateral border probably about 4-5 cm. Length of whole piece 5 cm. The first phalanx of the first left toe. The bone has apparently not suffered from pressure. Length 5-8 cm., proximal breadth 3-5 cm., proximal thickness 140 Annals of the Transvaal Museum 3-2 cm., distal posterior breadth 2-8 cm., distal anterior breadth i-6 cm., distal thickness 2-4 cm. The contour of the proximal articulation surface is trape- zoidal. The posterior border is much higher than the anterior one. The lateral side of the articulation surface is deeper than the medial side. The distal end is greatly twisted to the medial side. The contour of the distal end is also trapezoidal. It is not symmetrical. The articulation surface passes higher upwards posteriorly than anteriorly and its medial side projects further back- wards. The collateral pits are large. Another phalanx is the first of the second right toe. It is a first phalanx because of its evenly concave proximal articulation surface. It is one of the right side because the lateral portion of its distal articulation surface projects further downwards. It cannot be the first phalanx of the first toe, because of the pit on the distal end of the anterior surface and because of the depth of this pit and the thickness of the proximal end as compared with the proximal breadth. I take it to be the first phalanx of the second toe. Length 5-5 cm., proximal breadth 3-7 cm., proximal thickness 3*5 cm., distal posterior breadth 3-4 cm., distal anterior breadth 2-4 cm. and distal thickness 2-5 cm. The axis of the practically cylindrical proximal articulation surface is not parallel with the hinder border of this surface, but converges with it towards the lateral side. This means that the bone is slightly turned to the medial side, when it occupies its natural position. The medial portion of the pulley-shaped distal articulation surface projects further backwards than the lateral portion. Results. The main results from the preceding work are here enumerated : 1. Two new genera of the Theropoda have been described. 2. Another form has been doubtfully referred to a very little known species. 3. A redescription is given of Gryponyx transvaalensis Broom. 4. A remarkable difference has been found between the pubic neck of the Plateosaurid Eucnemesaurus and that of the Anchisaurid Dromicosaurus. It could not be ascertained whether this difference is of family value. 5. The femur of Eucnemesaurus disproves the generality of the rule that Theropod limb-bones are hollow. Perhaps this rule should be restricted to Anchisaurids. 6. v. Huene draws attention to the peculiar shape of the distal end of the tibia of Euskelesaurus and of Gresslyosaurus. Both these tibiae differ from all other Theropoda . It has now been shown that the published figures of the tibia of Gresslyosaurus agree with the tibia of all other Theropoda, but disagree with the accompanying description. 7. The Anchisaurid Dromicosaurus is shown to have distally coalesced pubes, a fact hitherto regarded typical of the Plateosauridae. 8. A great difference is shown to exist between the ischium of the Anchi- saurid Dromicosaurus and that of the Plateosaurid Teratosaurus. Could this difference be of family value? 9. Another great difference is pointed out between the Anchisaurid Dromicosaurus and Plateosaurus Quenstedti. Could this also be of family value? 10. What has been accepted to be the left lower arm of European Plateo- saurids is shown to be really the right lower arm. PLATES XI to XXIII PLATE XI. Eucnemesaurus fortis. Fig. i. „ 2. „ 3- » 4- Lateral view of left tibia, x \ . Posterior view of left tibia, x | . Medial view of upper end of left pubis, x Anterior view of upper end of left pubis. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XI PLATE XII. FAicnemesauYUs fortis. Fig. i . Lateral view of upper half of left femur. The upper ends of the trochanter major and the trochanter quartus are well visible. Slightly more than x ,, 2. View of polished lower end of the femur of fig. i. Natural size. ,, 3. Right side view of centrum of dorsal vertebra, x L ,, 4. Upper view of centrum of dorsal vertebra, x Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XII PLATE XIII. Eucnemesaunis fortis. 1. Left side view of caudal vertebrae. About J nat. size. Dromicosaurus gracilis. 2. Anterior view of left femur x The foramen nutritivum is well shown above the middle of the bone. 3. Lateral view of left femur, x 4. Posterior view of left femur, x 5. Postero-lateral view of proximal end of left radius. Slightly more than x 6. Proximal articulation surface of left radius. Nat. size. The anterior end is at the left-hand side 7. Lateral view of the distal end of the right metatarsale III. Slightly more than x 8. Lateral view of the distal end of the right metatarsale I Slightly more than x Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XIII PLATE XIV. Dromicosaurus gracilis. Fig. i. Anterior view of right tibia. Slightly more than x ,, 2. Lateral view of right tibia. Slightly more than x „ 3. Posterior view of right tibia. Slightly more than x ^ . The head is very much foreshortened in this figure, but its correct shape and position will be clear after comparison with figs. 1 and 2. ,, 4. Anterior view of the distal end of the right metatarsale III. x ^ . ,, 5. Lateral view of the proximal end of the left metatarsale III. x ., 6. Anterior view of the distal half of the humerus. Slightly more than x Annals Transvaal Mus . Vol. VII, Pt. 2 Plate XIV ! 5 6 PLATE XV. Dromicosaurus gracilis. Fig. i. Lateral view of right fibula, x L ,, 2. Anterior view of right fibula. Slightly more than x ,, 3. Medial view of right fibula, x ,, 4. Posterior view of the ischia, as preserved, x ,, 5. Lateral view of distal end of right ischium, x ,, 6. Medial view of proximal end of left ischium, x The articu- latio iliaca is above and the articulatio pubica at the lower right-hand border. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XV PLATE XVI. Dromicosaurus gracilis. i a. Left side view of two anterior caudal vertebrae. Slightly more than x A. The distal ends are to the right, i b. Left side view of an anterior caudal vertebra, posterior to those of fig. i a. Slightly more than x ic. View of the lower surface of an anterior caudal vertebra, posterior to that of fig. i b. Slightly more than x 2. Medial view of the proximal end of the left pubis, x 3. Left side view of the third neck vertebra. Slightly less than x 1. 4. Anterior view of the remains of the right pubis, x 5. Anterior view of the remains of the left pubis, x J . The pubes are placed with their medial sides facing each other. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XVI PLATE XVII. Massospondylus Browni. 1. Lateral view of right scapula and coracoid, x 2. Lateral view of the left humerus. Slightly more than x The circular depression on the posterior surface of the crista radialis is well visible. 3. Anterior view of the left humerus, x 4. Lateral view of left clawphalanx I. Nat. size. 5. Anterior view of first carpale of the left hand. Nat. Size. 6. Anterior view of second carpale Nat. size. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XVII PLATE XVIII. Massospondylu s Browni. Fig. i. Antero- medial view of left radius, x f. ,, 2. Antero-medial view of left ulna. Slightly more than x f . ,, 3. Antero-lateral view of left radius. Slightly less than x §. ,, 4. Postero-medial view of left ulna, x f . ,, 5. Upper view of left hand. The reconstruction is meant to show the hand at rest. Slightly more than x f . Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XVIII PLATE XIX. Massospcndylus Browni. Fig. i. Supero- lateral view of left hand. ' x f. 2. Lateral view of left ileum, x ,, 3. Anterior view of right pubis, x ,, 4. Lateral view of right ischium. Slightly more than x -J. Annals Transvaal Mus. Vol. VII , Pt. 2 Plate XIX PLATE XX. Massospondylus Browni. Fig. i. Anterior view of left femur, x ,, 2. Lateral view of left femur. Slightly more than x ,, 3. Medial view of right fibula, x ,, 4. Anterior view of right tibia, x ^ . ,, 5. Left side view of the sixth to the tenth caudal vertebrae, x J. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XX PLATE XXI. Massospondylus Browm. Fig. i. Upper view of right foot. The reconstruction is meant to show the foot at rest. Slightly more than x ,, 2. Left side view of ioth dorsal vertebra. Slightly more than nat size. ,, 3. Left side view of the five last neck vertebrae, x £. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XXI PLATE XXII. Massospondylus Browni. Superolateral view of right foot, xo-53- j j Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XXII PLATE XXIII. Dinosaur Vertebra. Fig. i. Left side view. Slightly more than x ,, 2. Posterior view. Slightly more than x „ 3. Ventral view, x The right side of the bone is on the right side of the figure. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XXIII Annals of the Transvaal Museum 141 In conclusion I beg to thank Mr H. Walker of St Fort and Mr W. H. Moore of Slabberts for their generosity; and Mr S. H. Haughton of Cape Town for valuable information. My thanks are also due to the Director of the South African Museum, Cape Town, for the loan of v. Huene’s “ Die Dinosaurier der europaischen Triasformation,” and, last but not least, to its author, for without the assistance of his great work I would not have been able to describe our material. Literature. The following literature has been referred to : 1. K. A. Zittel. Handbuch der Palaeontologie, Bd. hi. Leipzig, 1890. 2. H. G. Seeley. On the Type of the Genus M assospondylus and on some Verte- brae and Limb-bones of M. (?) Browni. Ann. Mag. Nat. Hist. vol. xv. Ser. 6, pp. 102-125. London, 1895. 3. K. A. Zittel. Text-Book of Palaeontology. Translated by C. R. Eastman. London, 1902. 4. F. v. Huene. Ueber die Dinosaurier der aussereuropaeischen Trias. Geol. u. pal. Abh., herausg. v. E. Koken, N. F. Bd. vm. H. 2. Jena, 1906. 5. F. v. Huene. Die Dinosaurier der europaischen Triasformation mit Beruck- sichtigung der aussereuropaischen Vorkommnisse. Geol. u. pal. Abh., herausg. v. E. Koken, Suppl.-Bd. 1. Jena, 1907-1908. 6. K. A. v. Zittel. Grundziige der Palaontologie. Neubearbeitet von F. Broili, E. Koken, M. Schlosser. ii. Abteilung. Vertebrata. Munchen und Berlin, 1911. 7. R. Broom. On the Dinosaurs of the Stormberg, South Africa. Ann. S. Afr. Mus. vol. vii. Pt. iv. pp. 291-308, Pis. xiv-xvii. London, 1911. 8. R. Broom. On the Remains of a Theropodous Dinosaur from the Northern Transvaal. Trans. Geol. Soc. S. A. vol. xiv. 1911, pp. 82-83, LI- xm- figs. 3> 4- Johannesburg, 1912. 9. R. Broom. Catalogue of Types and Figured specimens of Fossil Vertebrates in the American Museum of Natural History. II. Permian, Triassic and Jurassic Reptiles of South Africa. Bull. Amer. Mus. Nat. Hist. vol. xxv. Pt. 11. New York, Jan. 4, 1915. 10 142 Annals of the Transvaal Museum DESCRIPTION OF SOME CRETACEOUS AMMONITES FROM PONDOLAND By Dr E. C. N. van Hoepen, M.I. With three plates. The ammonites here described were purchased with other fossils from Mr J. Venter, who collected them on the coast of Pondoland, near the mouth of the Umzamba river. Phylloceras umzambiense n.sp. PI. XXIV, figs, i — 3. The whorls of the discoidal shell are strongly involute and higher than broad. The greatest thickness of the whorl lies internal to the middle of the flanks. The external surface is broadly convex and the sides are slightly flattened. The umbilicus is narrow and deep, the umbilical surface moderately inclined. The ornamentation consists of numerous, crowded, thin, wire-like ribs. These ribs commence deep down in the umbilicus and, on passing on to the flanks, swerve forwards; from near the midde of the flanks to the periphery they are practically straight and directed slightly backwards. The outer end of the ribs may be slightly stronger than the inner end, but on one portion of the shell the ribs are stronger on the internal half of the flank than on the outer. Short ribs, starting at or near the middle of the flanks and running across the periphery, are regularly intercalated between the long ones. Some- times, though very exceptionally, these ribs start nearer to the umbilicus. They all run practically straight to the periphery, in the same direction as the long ribs. The surface of the shell is weakly corrugated near the umbilicus. The corrugations are low and short, disappearing before reaching the middle of the flanks. They also disappear on the umbilical surface. Their curvature is the same as of the ribs in this region. Their breadth is such that they bear from four to six ribs. The lobe-line is phylloid. The anterior terminations of the saddles are broad and well rounded. This fact, however, is only seen well through the transparent shell; if this be removed with acid the lobe-line seems to suffer easily. In general the lobe-line agrees very well with that of Phylloceras For- besianum d’Orb sp. The siphonal saddle, however, is not so sharply pointed as in that species. The external lobe has nearly the same length as the first lateral lobe. The external saddle and the first lateral saddle are both sym- metrically divided by a secondary lobe. The second lateral saddle shows three anterior terminations. The auxiliary saddles, six in number, become rapidly smaller. Measurements : Diameter ... ... ... ... 44mm. (1) Height of last whorl ... ... ... 27 ,, (o*6i) Thickness of last whorl: actual 16 mm. was probably 17 ,, (0-39) Width of umbilicus ... ... ... 2 ,, (0-05) Superficially there is some resemblance between this form and Phylloceras Nera Forbes sp. The Indian species, however, has a series of radial sulci, around the umbilicus, which are apparently absent in our form. The radial corrugations of our form are apparently absent in P. Nera. The new species is also relatively thicker than P. Nera. There is a great difference between the Annals of the Transvaal Museum 143 lobe-line of the new form and that of P. Velledae and allied forms. In the Indian forms the external lobe is very much shorter than the first lateral lobe, whereas in the Pondoland ammonite the two lobes are nearly of the same size. There is close agreement between the lobe-line of our form and that of Phylloceras sp. mentioned by Woods from Pondoland (“The Cretaceous Fauna of Pondoland," Ann. S. Afr. Mus. vol. iv. Pt. vn. PI. XLI, fig. 4). There is, however, difference in the shape and the ribbing of the shell. There is also a great resemblance between the suture-line of P. Roger si Kitchin and that of our form. The shells differ, however, in shape and ornamentation. The specimen is imperfect and wholly septate. One side of the last whorl has weathered away. Gaudryceras amapondense n.sp. PI. XXIV, figs. 4, 5. Shell discoidal, compressed. Sides of whorl slightly convex and convergent towards the rounded periphery. Last whorl rather rapidly increasing. The umbilicus is large, the umbilical surface fairly steep and the umbilical margin strongly convex. Whorls higher than broad and covering about half of the flanks of the preceding whorl. Whorls ornamented with low ribs, which are narrow within the umbilicus and become broader towards the periphery. Between these other ribs are intercalated, which are only faintly visible at the umbilicus. Ribs of unequal breadth and separated by unequal distances. Both ribs and interspaces are covered by numerous fine ribs. The last whorl shows four ribs, at a distance of about 90 degrees from each other, which are much broader and thicker than those in their immediate vicinity. One of these ribs had to be removed in studying the suture-line and was seen to correspond with a constriction on the internal cast. The last of these ribs is immediately preceded by three others, which are also broad and thick and separated by broad interspaces. The umbilical portion of all the ribs is concave forwards, on the sides they are convex forwards and near the ventral border they are again concave forwards. Their ventral portion was not seen. Inner whorls covered with finer ribs; here and there is a coarser one followed by a deep groove. The suture-line has not been seen completely. That which is visible con- sists of a portion of the first lateral lobe, both lateral saddles, one auxiliary saddle and a portion of another. The remaining space between the visible portion of the last saddle and the umbilical suture is slightly broader than the breadth of this portion. The auxiliary lobes and saddles become smaller towards the umbilical suture and run slightly backwards. Further detail may be gathered from the figures. Measurements : Diameter ... ... ... ... 109 mm. (1} Height of last whorl ... ... ... 47 ,, (0*43) Thickness of last whorl ... ... perhaps 41 ,, (0-38) Height of penultimate whorl ... about 20 ,, (0-18) Thickness of penultimate whorl ... ... 9 ,, (0-08) Diameter of umbilicus ... ... ... 28 ,, (0-26) As the thickness over the longest diameter cannot be accurately given, the three first measurements are again taken at about 90 degrees back from the anterior end. Diameter ... ... ... ... ... 88 mm. (1) Height of last whorl ... ... ... 38 ,, (0-43 Thickness of last whorl ... ... ... 29 ,, (0*33 144 Annals of the Transvaal Museum The specimen is entirely septate. No species is known to me with which the new form could be confounded. Remarkable in the species, although by no means singular, is that the; breadth of the last whorl is relatively greater than that of the preceding one, and further the extraordinary height of the penultimate whorl with regard to its breadth. Tetragonites teres n.sp. PI. XXV, figs. 1, 2. Shell discoid al, umbilicated, with a few rapidly increasing whorls. Whorls higher than broad. Greatest thickness at the umbilical margin. Sides of whorl flat, becoming rounded towards the periphery. External surface strongly convex, rounding off gradually into the sides. Umbilical surface very steep. Inclusion two-thirds. The surface of the shell is smooth and covered with growth-lines, which are only visible with a magnifying glass. Two faint constrictions are visible, having the same shape as those of Tetragonites epi- gonum. Lobe-line very imperfectly known; as far as visible showing the general type of Tetragonites. Measurements : Diameter 69 mm. (1) At diam. of 57 mm- (1) Height of last whorl 39 , . (°'57) 3i - (o-54) Thickness of last whorl p 26 ,, (0-46) Diameter of umbilicus 14 , , (o-20) There is only one specimen upon which the above description has been based. About a quarter of its last whorl is probably body-chamber. The specimen has suffered somewhat from lateral compression. A portion of the last whorl, however, has not been compressed at all, and in this region the measurements on the diameter of 57 mm. have been taken. The new species differs from Tetragonites Timotheanum Mayor sp. and from T. epigonum Kossmat by the fact that its whorls are higher than broad ; in the two species mentioned they are broader than high. T. Timotheanum is also more involute and its whorls have an angular section. The section of the whorls of T. epigonum is also more trapezoidal than that of T. teres. The height of the last whorl is greater than half the diameter of the shell in the new species, whereas in T. Timotheanum it is equal to or, especially in T. epi- gonum and young specimens of T. Timotheanum, less than half this diameter. Holcodiscus Faku n.sp. PI. XXV, figs. 3, 4. PI. XXVI, figs. 1, 2. There are two specimens which are regarded as belonging here. The following description has been taken from the smaller of the two. Shell discoidal. Whorls higher than broad. Greatest thickness at the umbilical edge. Sides of whorl nearly flat near the umbilical margin, becoming more and more convex towards the periphery. External surface strongly convex. Umbilical surface low and very steep. Involution moderate, between one-half and two-thirds of the preceding whorl being embraced. Umbilical surface smooth. Sides of whorl covered with numerous, high and narrow ribs. All ribs pass with undiminished strength over the outer surface. The ribs emanate partly from strong, sharp, radially elongated tubercles, which have a flat and steep anterior surface and which stand on the extreme umbilical margin. The umbilical edge of these tubercles passes for a short distance on to the umbilical surface and stands practically perpen- dicular on the lateral edge of the protuberances. The ribs emanate from these tubercles in an irregular way. Sometimes the edge of one rib is continuous with the edge of the tubercle, while another rib emanates from its side. In other instances the edges of two ribs are continuous with the edge of the Annals of the Transvaal Museum 145 tubercle. Sometimes the edge of the tubercle is continuous with one rib, while another rib begins at each of its sides. Again, one rib is continuous with the edge of the tubercle, while another emanates from one side and two from the other. Sometimes the edge of the tubercle ends in a groove between two ribs. Now and then one of the ribs starts independently of any tubercle at the extreme edge of the umbilicus. Now and then, by no means at regular intervals, a short rib is intercalated between two others; these short ribs start at about the middle of the flank and behave further as the others. At irregular intervals the whorl is traversed obliquely by a deep groove, which is bordered behind by a thick ridge and in front by a narrow ridge. Its course is sinuous, being slightly convex forwards near its inner end, concave forwards near its middle and convex forwards at its outer end and on the external surface. The ridge in front of the groove starts on the umbilical surface, just inside the umbilical edge and passes on to the flank without forming a tubercle. It is not in touch with any other rib. The rib behind the groove starts at an umbilical tubercle and either bifurcates therefrom with another rib, or is the only rib continuous with the lateral edge of the tubercle, or emanates from the anterior surface of a tubercle, from which another rib starts in continuous line with its lateral edge and still another from its posterior side. This rib behind the groove cuts off two or three other ribs. As it crosses the external surface it becomes thicker and in the last two sulci of the larger specimen very much so. There are six of these grooves on the last whorl. The course of the other ribs is somewhat irregular. Many of them are very slightly convex forwards near the middle of the flank. A few are straight in this region. They mostly cross the external surface with scarcely any convexity forwards. In some of those near the sulci, however, the forward inflexion on the external surface is greater. The spaces between the ribs are much broader than the ribs and also broader than the nearest sulcus. On the end of the last whorl, which possibly partly represents the posterior end of the body-chamber, the interspaces become much broader and the ribs thicker. In the second specimen, of which apparently a large part of the body-chamber has been preserved, the ribs are very thick on this portion and the interspaces very broad. The lobe-line resembles most that of Holcodiscus. The external saddle and both lateral saddles are symmetrically divided by a secondary lobe. The external saddle is slender and longer than the others. The first lateral lobe is trifid and longer than the ventral lobe. The second lateral saddle does not touch the umbilical margin. On the umbilical surface the lobe-line runs backwards and forms three very small saddles. Measurements : Small spec. Large spec. Diameter 64 mm. (1) 62 mm. (1) Height of last whorl 27 „ (0-42) 26 ,, (0*42) Thickness of last whorl 21 ,, (°*33) 22 ,, (o-35) Height of penult, whorl 16 ,, 19 ,, (°'3°) Thickness of penult, whorl 13 » 16 ,, (0-26) Diameter of umbilicus 18 ,, (0-28) 18 ,, (0-29) The end of the last whorl of the type (small specimen) is damaged on one side and its thickness cannot therefore be accurately given. 21 mm. is its approximate dimension. The sides of the last whorl of the large specimen are also damaged and its thickness is therefore also given approximately. The measurements of the penultimate whorl of the type are not comparable with those of the last whorl, because they had to be taken on another diameter. 146 Annals of the Transvaal Museum The actual diameter of the second specimen is 82 mm. As the other measure- ments could not be taken on this diameter it was thought better to take them over that of 62 mm., whereby they would all be on one line. The new form shows great resemblance with species of the group Holco- discus Aemilianus Stol. It differs from H. Kandi Stol. sp. by being more involute, by the different arrangement of the ribs and their comparative straightness. H. madrasinus Stol. sp. has a last whorl, which is twice as high as broad. It is also more involute and its constrictions are more numerous. In H. karapadensis Kossmat the ribs are flattened on the flanks and on the periphery, while the sulci are directed much more forwards. H. buddhaicus Kossmat shows great resemblance in the ornamentation, but differs in the fact that its whorls are broader than high. Named after Faku, a former paramount chief of the Pondo tribe. Holco discus africanus n.sp. PI. XXVI, figs. 3 — 5. Shell discoid al. Last whorl higher than broad . Greatest thickness of whorls at the umbilical edge. Sides of whorl nearly flat near the umbilical margin, becoming more and more convex towards the periphery. External surface strongly convex. Umbilical surface low and very steep. Involution moderate, about one-half of the preceding whorl being embraced. Umbilical surface smooth. Sides of whorl covered with numerous narrow, rounded ribs, which are strong on the sides but ver}^ faint at the periphery. Most of the ribs emanate from tubercles on the extreme umbilical margin. The tubercles have mostly a more or less radially directed edge; the ribs emanate from them by twos, while a rib originates at the side of the tubercle in front and behind of this pair. Practically, therefore, there are four ribs to one tubercle. Near an old mouth groove this arrangement may become slightly irregular, through the presence of an extra rib. Ah the ribs have the same relative length. At irregular intervals the whole whorl is traversed by a deep groove, which is bordered in front by a thick ridge. Near the periphery the hinder border of the groove is formed by an equally strong rib, but on the flanks this rib is very much weaker. The course of the groove is sinuous; at the umbilical edge it is concave forwards, near the middle of the flank it is convex forwards and past this middle it is again concave forwards. Its external end makes a strong anterior sweep towards the periphery. On the periphery it is strongly convex forwards. The anterior rib does not form a tubercle, although it is prominent on the umbilical edge. It is not in touch with any other rib. That portion of the posterior rib, which is actually rib, starts at or near the middle of the flank; the first rib behind this begins at the posterior edge of the groove, about midway between the middle of the flank and the umbilical edge. The second rib behind it starts on the anterior surface of a tubercle, which stands on the edge of the groove. There are four of these constrictions on the last whorl. The course of the other ribs is very regular. They have all got a stretched S -shape, their inner half being very slightly convex and their outer half concave forwards. As far as they are visible on the periphery they are convex forwards. The spaces . between the ribs are broader than the ribs and on the middle of the flank only slightly broader than the nearest sulcus in this region. The lobe-line resembles most that of Holcodiscus karapadensis and H. buddhaicus. The external saddle and both lateral saddles are symmetrically divided by a secondary lobe. The external saddle is slender and longer than the others. The first lateral lobe is trifid and longer than the ventral lobe. The second lateral saddle does not touch the umbilical margin. The first auxiliary Annals of the Transvaal Museum 147 lobe lies external to the umbilical margin. It is very short and its posterior end lies far in advance of the posterior end of the second lateral lobe. On the umbilical surface the lobe-line runs as far backwards as the posterior end of the second lateral lobe. approximately about about 44 mm 17 „ 12 ,, 8 ,, 8*5 mm 13-5 „ (1) (o-39) (0-27) (0*18) (0-19) (o-3i) Measurements : Diameter Height of last whorl Thickness of last whorl Height of penultimate whorl Thickness of penultimate whorl Diameter of umbilicus This is another form belonging to the group of Holcodiscus Aemilianus Stol. sp. It differs from H. Kandi Stol. sp. by having the ribs flattened on the periphery and by the fact that all the ribs start at the umbilicus. The constrictions have also a different shape. In H. madrasinus Stol. sp. the ribs pass with undiminished strength across the periphery, while the constrictions are more numerous and cut off more ribs. There is great resemblance between our form and H. karapadensis Kossmat. In the Indian ammonite, however, the ribs are not conspicuous on the inner portion of the flank and the sulci are directed more forwards, so that more ribs are cut off. The shape of the sulci is also different. The height of the whorl augments quicker in our form than in H. karapadensis. In H. buddhaicus the whorls are broader than high. One side of the last whorl shows a portion of the last mouth edge. If this edge is identified correctly, the body-chamber occupied slightly more than half a whorl. PLATES XXIV to XXVI PLATE XXIV. Phylloceras umzambiense. Fig. i. Lateral view, x 1*87. ,, 2. Half outline of transverse section at greatest diameter. Natural size. ,, 3. Suture. Natural size. Gaudryceras amapondense. ,, 4. Lateral view. Natural size. ,, 5. Half outline of transverse section at diameter of 99 mm. Natural size. PLATE XXV. Tetragonites teres. Fig. i. Lateral view. Slightly less than natural size. ,, 2. Outline of transverse section through last whorl, where it has not suffered from pressure, at a diameter of 57 mm. Holcodiscus Faku. ,, 3. Lateral view of type. Natural size. ,, 4. Outline of transverse section at a diameter of 56 mm. Natural size. Annals Transvaal Mus. Vol. VII, Pt. 2 Plate XXV PLATE XXVI. Holcodiscus Faku. Fig. i. Lateral view of large specimen. Natural size. ,, 2. Suture of large specimen. Natural size. Holcodiscus afvicanus. ,, 3. Lateral view. Natural size. ,, 4. Half outline of transverse section at diameter of 40 mm. Natural size. ,, 5. Suture. Natural size. Annals Transvaal Mus. Vol. VII , Pt. 2 Plate XXVI MEDEDELINGEN ANNALS OF THE VAN HET TRANSVAAL MUSEUM VOLUME VII PART 3 containing , . _ t On the South African Notodontidae , with Descrip- tions of Apparently New Genera and Species. By A. J. T. Janse. (With 14 plates.) Issued June 30 th> 1920 PRINTED AT THE UNIVERSITY PRESS CAMBRIDGE, ENGLAND 920 ANNALS MEDEDELINGEN OF THE VAN HET TRANSVAAL MUSEUM Vol. 7 Part 3 ON THE SOUTH AFRICAN NOTODONTIDAE WITH DESCRIPTIONS OF APPARENTLY NEW GENERA AND SPECIES As far as the sub-continent is concerned, the family Notodontidae has, on the whole, neither been extensively dealt with nor properly classified. Further, no attempt has yet been made to catalogue this group of South African moths, although the country is fairly rich in species. It has been my good fortune to secure and study the majority of these, and only three genera recorded from S. Africa have not been before me. The most comprehensive account of the family is to be found in Professor Packard’s Monograph of the Bombycine Moths of America north of Mexico. This work deals only with North American genera of which few indeed occur in S. Africa. It has nevertheless been of great assistance to me although I have not been able to apply the principles of sub-family grouping therein annunci- ated owing to slight acquaintance with local larvae. In the future, therefore, when these have become better known, especially the larvae of the first in star, some rearrangement of the genera may become necessary. In the task which I have set myself, I have also found the Moths of India, vol. 1. by Sir George Hampson most useful although not more than ten Indian genera occur in South Africa. Further, a paper on the American Notodontidae by Schaus which appeared in the Trans. Ent. Soc. Bond. 1901 (pp. 257-343), and several papers on European members of this family have also served a useful purpose. However, the generosity of Sir George Hampson in supplying me with notes on genera, which have no doubt led to the avoidance of many an error, has been my main assistance and I take this early opportunity of expressing my deep sense of gratitude to him. As usual, the conventional but entirely artificial northern boundary of South Africa has been adopted; i.e. the southern banks of the Zambesi and Quinene rivers. The material examined is typical of the whole area south of this line of demarcation. Most of the specimens came from Natal, the Transvaal or Southern By A. J. T. Janse. With 14 plates. / \ r Annals of the Transvaal Museum 150 Rhodesia, the last yielding several new and peculiar forms. I have had access to nearly all the available specimens in our local Museums and private col- lections; among the latter I would mention those of Messrs E. E. Platt, E. L. Clark and Father J. A. O’Neil. So far as my knowledge of the moths goes, the distribution of the Noto- dontidae in S. Africa is interesting inasmuch as scarcely any species peculiar to the Cape Province have come under my notice. This does not obtain with the Lymantriadae , of which certain genera were rather well represented in that region. Although it is by no means so rich in Heterocera as the Eastern region, I cannot help thinking, that a more systematic survey would show more N otodontidae in the Cape Province than we know of at present. It is not unlikely that some of the insects herein described as new species have been previously described from Central Africa, because the references at my disposal do not include all the literature of the N otodontidae of Africa; however, every possible precaution has been taken to guard against duplica- tion. The references given are those which I have used, but all synonyms have not necessarily been given for each case. The measurement of the span has been arrived at by doubling that taken from the middle of the thorax to the tip of the fore wing. Unless stated to the contrary, the point at which an inner vein branches away from an outer is a proportion of the length of the outer, as measured from the cell to the margin In many instances the interrelationship of the various genera has not been very clear to me. There are gaps which may yet be filled by new forms or which were once occupied by some now extinct. Packard has defined with more or less success seven sub-families. For my own part I recognise eight groups, but it is questionable whether these have the value of sub-families. Group I. This group comprises the genera Scalmicauda and Ichthyura, of which the members are characterised by their hairy eyes, and their tendency to develop tufts on the thorax, peculiarities of no other Notodontids. Ichthyura is the more specialised genus, but I hardly think that it originated from Scalmicauda. Group II. The two genera Pectinophora and Lophopteryx compose this group of moths peculiar for the tufted inner margin of the fore wing. I take this group to have originated from an ancestral form common to it and to Antheua as its structure is rather more primitive than that of Antheua although similar. Group III. This group I also regard as having originated from the root common to groups II and V, although no trace can be found elsewhere of the spur on the fork of vein 1 b of fore wing, peculiar to this group and possibly a rather primitive character. This Cerura group has three genera: Pseudorethona, Cerura and Pararethona ; the spur on 1 h, the scaling of the legs and the branches of the antennae all indicate close affinity. Cerura has lost the median spur on the hind legs, and for this reason is considered specialised; but, on the other hand, it has a bar between the upper median, and vein 8 of the hind wing; this bar is missing in Pseudorethona. Pararethona may ultimately prove to be con- nected in one way or another with Chadisra but it certainly also exhibits relationship to the two other genera of the group. 14 Rigema Annals of the Transvaal Museum 151 n — 2 22 TseniopteryK 152 Annals of the Transvaal Museum Group IV1. This may be called the Desmeocraera group, comprising the genera Desmeo - craeva, Phyllaliodes, Hypophiala and Stauropus, the members being peculiar as the proboscis is absent or, when present, very weakly developed. Of these genera Desmeocraera is the largest South African genus and peculiar for (i) the absence of the areole, (2) for its well developed palpi, (3) for possess- ing spurs on the hind legs, (4) for the presence of all veins, (5) for the smallness of the third joint of the palpus. Stauropus originates directly from Desmeo- craera, having all the veins but only two spurs on the hind legs; in both genera the terminal part of the spurs is peculiar and the palpi are reduced in size and porrect. The remaining two genera Hypophiala and Phyllaliodes may have originated near Desmeocraera, but the position of Phyllaliodes in particular, is far from clear to me as veins 6 and 7, 8, 9 and 10 suggest that they arrived from a form having an areole in the fore wing. The palpi in both genera are however very much reduced. There exists in these two genera a remarkable likeness to the Striphnopterygidae in general appearance. Group V. This group should most probably include Antheua, from which Phalera branches off more or less directly as it shows a close affinity to that genus, being differentiated from it by (1) the more elongated fore wing; (2) the position of veins 6 and 7; (3) the presence of the bar in the hind wing; (4) the shorter and less upturned palpi; (5) the reduced pectination of the antennae; (6) the general pattern of the fore wing, the last differentiating it at once. Both Rigema and Leucophalera may have originated from Phalera. In Rigema the teeth of the spurs have been more developed, the palpi are more porrect and shorter, but the markings on the abdomen are as in Antheua and Phalera. Leucophalera shows a similar development of the spurs, but the rows of teeth are shorter, the bar in the hind wing is absent, in the fore wing veins 6 and 7 originate in a different way and the process of the fore tibia is also different from that of any of the previous genera. Group VI. This may be named the Chadisra group. It comprises Chadisra, Hoplitis, Galona and Amy ops. The genus Chadisra shows affinity to Antheua in the structure of the palpi and in the venation of the wings; but the hind wing has a bar, and the fore wing is more narrow and elongate with a corresponding decrease in surface of the hind wing ; these two last characters are found more or less in all genera of this group. There is also a tendency for the branches of the antennae to be reduced; these in the more generalised genera are rather well developed. The process of the fore tibia is of the same type as that of Antheua and the spurs, though longer, terminate in the same manner. I think two branches sprung from this genus; one being Hoplitis, which preserves the upturned palpi but has lost the areole through anastomosis, the other, com- 1 Groups IV, V and VI are in many ways closely allied and I regard them as originating from a form very similar to the present members of Antheua. Possibly they should be included with that genus so as to form one large group. Antheua is certainly the most generalised of all the remaining genera although containing less species than one would expect for such an old genus. Even nowadays Antheua displays a great tendency to vary structurally. It is peculiar, that the bar between the upper median and vein 8 of the hind wing is absent in Antheua, present in groups VI and VIII and absent again in group IV. Annals of the Transvaal Museum 153 prising Galona and Amyops which preserve the areole whilst the palpi become much reduced; in Amyops the median spurs of the hind legs also disappear. Group VII. This may be called the Scrancia group, comprising six genera. It may have originated somewhere near Chadisra , the similarity in structure and the ten- dency of the fore wing to be narrow, pointing to this. But here the hind wing is ample and remains usually so, while the rather long legs, especially of Scrancia, are extraordinary features in the Notodontidae. I think that Phycitimorpha is the oldest genus and from it two branches have developed: the “Scrancia” branch, in which the proboscis remained, and the “ Breyeria” branch with this organ lost to its three genera. From Scrancia, Taeniopteryx may have developed, judging by (1) the coalescence of the areole of the fore wing, (2) the stalking of vein 6 to the stalk of 7-10, (3) the obsolescent vein 5 in the hind wing; a puzzling feature is, however, the rather long process of the fore tibia which in Scrancia is short and broad. It is remarkable that in Breyeria this process is also longer than in Scrancia, even longer than in Phycitimorpha', further the members of Breyeria lack a proboscis, have reduced palpi and much broader fore wing than those of the Scrancia branch. The members of the genus Stenostaura are remarkable for (1) the stalking of veins 3 and 4 in both wings, (2) the absence of the median spurs on the hind legs, (3) the shorter piocess on the fore tibia, and (4) their palpi. In Eurystaura on the other hand the areole is lost by coalescence, vein 5 of the hind wing becomes very faint and the process of the fore tibia is much shorter than in Breyeria. Group VIII. This I call the Notoxantha group. It comprises as many as eleven genera, several of which have a more or less developed bar in the hind wing. The group originated somewhere near Antheua, Zana in particular closely resembling Antheua. Notoxantha shows several generalised characters but is probably not quite like the form from which the others must have come. Crambometra is peculiar for the well-developed branches of the antennae; the venation of the fore wing resembles Notoxantha, but in that of the hind wing vein 5 is reduced; the genus Campyloctys is closely allied to Crambometra as is clearly indicated by the antennae, but the venation of the fore wing is remarkable, as veins 8 and 9 have become separated, thus doing away with the areole and causing 7 and 8, 9 and 10 to be stalked; here also the palpi are more porrect. Zana may have originated directly from Notoxantha, the spurs are shorter, the bar of the hind wing is less distinct and the palpi are more reduced and porrect. Ramesa stands more or less alone by the peculiar development of tlie antennae, but the fore wing agrees in venation, but not in markings, with Notoxantha. Polienus and Prionocentrum come somewhere before Pydna, but the relation of both these genera is not quite clear to me; on the whole Priono- centrum is a rather peculiar Notodontid. The peculiar reduction in the Pydna branch is the disappearance of the areole through anastomosis of the greater part of vein 10 with veins 6, 7, 8, 9 (in P. rubrifascia, indeed, it can be seen how the areole disappeared) ; all other characters are remarkably like those of Notoxantha except that the frons is smooth. Polelassothys has (1) the palpi very much reduced, (2) only two spurs on the hind tibia, (3) vein 8 of hind wing coalescent with nearly the whole of the upper median; this last feature is particularly interesting. The genus Anaphe may have originated from Pydna, although the reduction of the spurs as between these two genera points to one or more intermediate 154 Annals of the Transvaal Museum genera. Epanaphe I consider to be directly derived from Anaphe, although the disappearance of vein 5 in both wings is a very sudden transition. ACKNOWLEDGEMENTS. The Transvaal Museum: for the most generous way in which the Director allowed me to publish this paper and Mrs C. J. Swierstra for assisting me with his material. The South African Museum: for the loan and presents of several Notodontids and for kindly copying some descriptions not in my library. The Natal and Durban Museums: for the loan of their whole material of this family. The Southern Rhodesia Museum : for the loan of specimens. Annals of the Transvaal Museum 155 Family NOTOD ONTID AE. General build noctuiform, seldom slender; fore wing 1 b often forming a fork at base, 1 c absent, vein 5 from middle or a little above middle of discocellulars ; hind wing with two internal veins, vein 5 from middle of discocellulars or absent, 8 free from base, then parallel with, or approximated to, upper median or connected with that vein by a bar, then obliquely to near apex. KEY TO THE GENERA. 1 a. Eyes hairy ............ 2 b. Eyes glabrous ........... 3 2 a. Fore wing with an areole ..... Scalmicauda p.158 b. Fore wing without an areole .... Ichthyura p. 161 3 a. Fore wing with one or more tufts of hairs and scales on inner margin at middle ........... 30 b. Fore wing without such a tuft ........ 4 4 a. Hind legs with four spurs ......... 10 b. Hind legs with two spurs only ........ 5 5 a. Fore wing with an areole ....... . 6 b. Fore wing without an areole . . ... . . 7 6 a. Areole long, about half the distance from upper angle to apex; abdo- men with lateral tufts of hair ..... Amyops p.206 b. Areole short, less than one-third the distance from upper angle to apex; no lateral tufts of hair on abdomen .... Gerura p. 170 7 a. Fore wing with all veins present ........ 8 b. Fore wing with at least one vein absent ...... 9 8 a. Antennae bipectinated for two-thirds their length; branches in over six times shaft; fore and hind wing with veins 3 and 4 separate; hind wing with vein 5 rather weak .... Stauropus p. 194 b. Antennae bipectinated for two-thirds of shaft, but pectination short, only two times shaft; hind wing with vein 5 weak; fore wing with veins 3 and 4 on a short stalk; hind wing with veins 3 and 4 on a stalk of over one-third of 3 or from a point . . Stenostaura p. 215 c. Antennae bipectinated till the tip .... Anaphe p.232 9 a. Fore wing with vein 8 absent; hind wing with all veins present Polelassothys p. 229 b. Fore wing with veins 5 and 9 absent; hind wing with vein 5 absent Epanaphe p. 234 10 a. Hind wing with vein 5 absent or very weak1 ..... 31 b. Hind wing with vein 5 present, at least distinct . . . . 11 11 a. Fore wing without an areole2 ........ 12 b. Fore wing with an areole3 ........ 15 1 In Antheua dimovpha vein 5 is very weak and the proboscis is in some specimens practically absent, but the other characters place this species provisionally in Antheua. 2 In some cases the areole is not completed by the connection between vein 10 and the stalk of 8-9. General characters will have to decide to which group it belongs. In Antheua some species are very inconsistent in this character, some specimens of the same species having and lacking the areole. Many specimens have in such a case to be examined. 3 In Chadisra and Antheua the areole may be absent in some species. 156 Annals of the Transvaal Museum 12 a. Fore wing with vein io free Hypophiala p. b. Fore wing with vein io stalked with 8, 9 . . . Phyllaliod.es p. c. Fore wing with vein 10 stalked with 7, 8, 9 13 a. Fore wing long and with vein 10 from before 7; outer margin very oblique; spurs on tibia short ...... Hoplitis p. b. Fore wing with vein 10 from beyond vein 7 ..... 14 a. Pectination of antennae in $ long, over 4 times shaft, but tip not pectinated; fore wing with vein 6 from base of stalk 7, 8, 9, 10 Desmeocraera p. b. Pectination of antennae of $ short, about two times shaft, but almost reaching to tip; 6 of fore wing remote from base of stalk 7, 8, 9, 10 . . . . . . . Pydna (rubritincta) p. 15 a. Fore wing long, outer margin very oblique; legs and spurs very long; areole of fore wing very broad, about as long as three times breadth ; pectination of antennae of $ short (two times shaft) and only up to about frd of shaft; antennae over half of costa in length; palpi long, upturned, second joint three times first joint, covered with short hairs ; hind wing with vein 8 parallel to upper median from base to half of upper median ...... Scrancia p. b. Fore wing long, outer margin erect; areole broad (about three times), vein 6 from frd of areole; 7 on a long stalk with 8, 9; hind wing with vein 5 very weak; pectination of antennae over three times shaft ........ Phycitimorpha p. c. Fore wing broad; legs moderate; areole over three times as long as broad; antennae with the pectination three times shaft or more in $ ; palpi rather short, densely hairy ...... 16 a. Fore wing with vein 6 remote from the areole, or if it is as in b the abdomen has distinct dorsal crests (as in Polienus nigrosparsa) ; hind wing with vein 3 from frd distance 2-4 ; 5 rather weak ; hairs of thorax produced into one small tuft ...... b . Fore wing with vein 6 from the areole or from base of areole, or if as in a the abdomen has no dorsal crests; hind wing with vein 3 from well beyond frd distance 2-4 ; hairs of. thorax not forming a tuft . 17 a. Areole very long, about half the distance from upper angle to apex; 10 from areole; 8 of hind wing not connected by a bar with the upper median .......... Galona p. b. Areole very short, about fth of distance upper angle to apex, 10 from stalk of 8, 9; 8 of hind wing connected with upper median by a bar .......... Polienus p. 18 a. Fore wing with vein 6 from end of areole, almost stalked with 7, 8; 5 from near areole, much above middle of discocellulars b. Fore wing with vein 6 from well before end of areole; 5 remote from areole, not much above middle of discocellulars .... 19 a. No spur towards inner margin on fork of vein 16 of fore wing; no tufts of scales on mid and hind tibiae; branches of antennae do not reach the tip and are about 10 times the shaft in the 31; palpi with 1st joint longer than 2nd joint; spurs with two rows of serrae Pararethona p. b. Fore wing with a spur towards inner margin from fork of 16 as in Cerura ; mid and hind tibia with tufts of scales ; antennae pectinated till tip and branches about four times shaft in <$; palpi with 1st and 2nd joints of equal length; spurs of tibia with two very short rows of slight serrae only .... Pseudorethona p. 20 a. Fore wing with vein 7 from before 10 or from equal distance from areole, or from beyond 10 and then at any rate from before end 192 192 13 203 14 183 229 210 208 16 17 18 205 223 19 20 173 168 Annals of the Transvaal Museum 157 of areole (as in A. simplex, where 7 comes from the stalk just beyond 10) ........... b. Fore wing with vein 7 from well beyond vein 10 21 a. Fore wing with vein 10 from the areole ...... b. Vein 10 stalked with 8, 9 22 a. Fore wing rather produced, outer margin somewhat incurved between apex and vein 4, termen rather strongly sinuated; branches of antennae in till a little over middle of shaft and a little longer than thickness of shaft . . • . . . . . Rigema b. Fore wing not produced at apex, outer margin straight; antennae in $ bipectinated till at least frd of shaft and pectination about two times shaft; outer margin but slightly sinuate or not at all Antheua 23 a. Fore wing with vein 10 very shortly stalked with 8, 9; areole extremely narrow ....... Pydna (rubrifascia) b. Fore wing with vein 10 on a stalk of about Jrd the distance of end of areole to apex; areole rather broad ...... 24 a. Apex of fore wing produced; fore wing very broad, termen well curved inwards below apex ..... Prionocentrum b. Apex of fore wing not produced, termen not curved inwards below apex ........... 25 a. Palpi obliquely upturned; hind wing with stalk of 6, 7 at least ^rd of 7 b. Palpi porrect; hind wing with stalk of 6, 7 very short, only about |th of 7 ......... Leucophalera 26 a. Frons with a horse-shoe-shaped prominence; antennae bipectinated till near the tip ; termen rather erect, not sinuate ; spurs rather thin and long ........ Notoxantha b. Frons smooth; pectination of antennae till a little beyond middle of shaft, or only fasciculated; termen oblique and somewhat sinuate; spurs rather short and thick ..... Phalera 27 a. Fore wing rather long and narrow; 7 stalked with stalk of 8, 9 . b. Fore wing broader and outer margin more erect (as in Antheua) ; 7 and stalk of 8, 9 from or from very near end of areole .... 28 a. Fore wing rather long, termen very oblique; a bar between upper median and vein 8 of hind wing ; hind wing with dark anal macula Ghadisra b. Fore wing broad; termen erect; no bar in hind wing; hind wing con- colorous at anal angle ...... Breyeria 29 a. Branches of antennae very long, about 8 times shaft, shaft curved and the ends of the branches forming nearly a straight line, extending till the tip ; mid-spurs of hind legs short (not reaching beyond end of tibia) ; tibia densely covered with very long hair ; areole of fore wing very long, reaching about half the distance from upper angle to apex ......... Galona b. Pectination of antennae only till fth of shaft, then the shaft becomes dentate; pectination about 3-4 times shaft, straight, not curved; hind legs with mid-spurs longer and more slender (reaching well beyond end of tibia) ; hairs on legs less dense ; areole moderate, reaching till about ^rd of distance from upper angle to apex . Zana c. Antennae unipectinate till half length of shaft, then dentate; pectina- tion short (not over two times shaft) ; spurs of hind legs long, reaching far beyond end of tibia, smaller spur about half of long spur ; areole long, length over half the distance of from upper angle to apex Ramesa 30 a. Areole short, about ^rd of vein 10; outer margin of fore wing almost 21 27 22 23 p. 197 P- I75 p. 229 24 p. 226 25 26 p. 199 p. 217 P- 195 28 29 p. 200 p. 213 p. 205 p. 221 p. 222 158 Annals of the Transvaal Museum straight; 6 remote from areole; palpi upturned, long, with 2nd joint nearly half the length of the palpus . . . Pectinophora p. 164 b. Areole long, about half length of vein 10; outer margin sinuate; 6 from areole; palpi short ..... Lophopteryx p. 166 31 a. Fore wing with an areole; hind wing with vein 5 present but very weak ............ 32 b. Fore wing without an areole; hind wing with vein 5 absent ... 33 32 a. Fore wing with the areole long, 6 times breadth, formed by 8 and 9 becoming shortly stalked beyond the areole; 6 remote from the areole; antennae bipectinate, branches very long, over 8 times shaft, curved downwards; palpi upturned, rather long; hind wing with 6 and 7 remote ..... Grambometra p. 218 b. Hind wing with vein 5 present, but very weak; fore wing with vein 6 from over frd of areole; 7 stalked with 8, 9; areole about three times longer than broad; antennae bipectinate, branches nearly four times shaft, straight; palpi obliquely upturned, just reaching vertex; frons rounded (see also 15 b.) . . . Phycitimorpha p.208 33 a. Fore wing with 7, 8, 9, 10 stalked and from upper angle; 6 from upper angle or stalked with 7, 8, 9, 10 . . . . . . . 34 b. Fore wing with 7 and 8 stalked for nearly half of 7, and from upper angle; 6 from before upper angle; 9 and 10 on a stalk of half of 10 and from before upper angle .... Campyloctys p. 219 34 a. Fore wing with vein 6 on a long stalk with 7, 8, 9, 10; hind wing with veins 3 and 4 on a stalk of ^rd of 3 . . . Taeniopteryx p. 212 b. Fore wing with vein 6 very shortly stalked with 7, 8, 9, 10, or from upper angle together with this stalk; hind wing with 3 and 4 remote ............ Eurystaura p. 216 Genus SCALMICAUDA. Scalmicauda Holl. Psyche, vi. p. 439 (1893). TyPe benga. Auriv. Entomologisk Tidskvift, p. 197 (1892). Auriv. Arkiv for Zoologi, Band 2, No. 4, p. 5 (1904). (PL I, figs. 1-5; PI. II, figs. 1-5.) Description from albicostata Hmpsn. $. Tongue short but present; palpi rather short, a little longer than length of eye, obliquely upturned, thickly scaled and with some long hair on first joint; first joint short, about one-third of palpus, curved; second joint cylindrical, straight, nearly two times first joint; third joint minute, completely hidden in hairs; eyes covered with hairs, rounded, rather large; antennae about half length of costa, usually curved, in $ bipectinate for about two-thirds, or as in o’neili bidentate for the same length and with a bundle of curved cilia at the end of each tooth; branches about three times shaft and ciliated for- wardly ; antennae in $ simple, ciliated ; fore tibia with a hollow curved process reaching till end of tibia; mid tibia with terminal spurs; hind tibia with four spurs which end in a curved point; femora with long moderate hairs, fore tibia with a thick covering of hairs, mid and hind tibiae sparingly covered with hairs ; tarsi with bristles on inner side, smoothly covered with hairs generally. Fore wing rather long and narrow; costa nearly straight, curved towards apex, which is roundly pointed; termen oblique, gently curved between veins 3 and 7; tornus well rounded; inner margin excurved at middle; vein 1 b faintly forked at base ; 2 from f th lower median ; 3 from nearly fth 2 to 4 ; 4 from lower angle ; 5 from above middle of discocellulars which are outwardly oblique ; Annals of the Transvaal Museum i59 two short veinlets in the cell; 6 from about middle of areole, which is nearly as long as half vein 10; 7, 8, 9 stalked and from end of areole; 7 from stalk at -frd of whole stalk, which is as long as f distance end of areole to apex; free part of vein 10 from end of areole; 1 1 from upper median at frd, parallel to 12 ; 12 parallel to costa. Hind wing sub-triangular; costa well curved; apex well rounded; termen oblique and well curved ; tornus rounded ; inner margin gently curved ; 1 a and b somewhat curved, long; 2 from lower median at fth; 3 from fth distance 2 to 4 ; 4 from lower angle and curved; 5 from above f discocellulars which are somewhat oblique inwardly and have a forked veinlet in the cell; 6 and 7 on a stalk of frd 7 and from upper angle ; 8 curved upwards at base, then down- wards and approximated to upper median, then curved upwards; a bar to vein 8 from just before middle of upper median. The hairs of the thorax have a tendency to produce a tuft, which is very long and thin in heterogyna and practically absent in albicostata; the abdomen has a brush of long hairs, which is however absent in albicostata. The description given by Holland is altogether insufficient to fix the genus and as the type of the genus is unknown to me, it may be that the characters given above do not apply to all species. Aurivillius in Arkiv for Zoologi also remarks, that Holland’s description is insufficient, and adds many useful characters which refer to fuscinota, all of which are found on the species I place in this genus. Hampson in A .M.N.H. 8. v. p. 492 places Aurivillius’ species for certain in Scalmicauda, and in addition, gave me very useful information about the genus, also mentioning the hairy eyes, so that I have no doubt that my species are rightly placed here. It is to be regretted, that generic descriptions are sometimes incomplete to such an extent, that it is impossible to recognise the genus with any certainty if the actual type species is not at hand. The original description should, I think, mention every character more or less readily observable and of dis- tinctive value. This not only applies to genera but also to species. The genus seems to be confined to Africa ; three species have been described from South Africa to which I add a fourth. They may be tabulated as follows : la. A long, narrow thoracic tuft of hairs .... heterogyna b. No thoracic tuft at all, or if present not long and narrow ... 2 2 a. White points at the angle of cell of fore wing; a black point in cell near base; white points at base of cilia .... griseitincta b. No white points on fore wing and no black point in cell near base . 3 3 a. Costa narrowly white, fore wing maroon-red, no orbicular, reniform or macula below lower median ; hind wing in both sexes white ; antennae of bipectinated for frd ..... albicostata b. Costa of fore wing in $ broadly white, in $ this white is reduced to a triangular macula at end of post median line; a brown round orbicular, reniform and round macula below lower median and before vein 2; cilia tipped with white; antennae of $ bi-serrate for frd ........... o’neili. Scalmicauda Heterogyna. (PL I, fig. 2.) Scalmicauda heterogyna Hmpsn. A. M.N.H. 8. v. p. 490 (1910). I have a $ and a and have seen another $ which is in Mr Clark’s collection, •all in rather broken condition, but distinctly marked. They differ from the description in a few points: the brown rings around the orbicular and reniform i6o Annals of the Transvaal Museum are continuous in one specimen, and the hind wings are whitish-yellow, not white. The process of the fore tibia is in this species similar to that of albicostata, but is less twisted and curved inwardly at the tip, while the covering of hairs is. much more dense and thicker than in albicostata. Hab. Gillets (Natal), 15. 2/17; , Durban, 5. 5. ’13, both in collection Janse; $ from Durban in collection Clark. SCALMICAUDA GrISEITINCTA, Scalmicauda gristeitincta Hmpsn. A.M.N.H. 8. v. p. 492 (1910). I have not met with this species yet. Scalmicauda Albicostata. (Pi. t, fig. 1; Pi. 11, figs. 1-5.) Scalmicauda albicostata Hmpsn. A.M.N.H. 8. v. p. 493 (1910). My specimens agree well with the description, but the white edging on the costa of fore wing is very narrow; also the colouring of the antemedial line is more yellow and not whitish and in the $ the veins of the fore wing are ochreous in colour, a character not mentioned in the description. In a letter to me, Sir George Hampson mentions these characters also from his specimens, so there can be no doubt about my specimens belonging here. Hab. Umvuma (S. Rhodesia), 25. 12 ’17; Salisbury (29. 2. ’17; $ and $ bred by Father O’Neil1.) One of these rfs is very dark in colour and the white costa is hardly present. My ^s have the antennae bipectinated for frd, the apical third being simple. The costa of the fore wing in the $ is less arched than in the $ and the antemedial and postmedial lines are more defined in the $ than in the . The thoracic tuft is in this species only indicated. Scalmicauda O’Neili spec. nov. (PI. XIII, figs. 1-3; PI. I, figs. 4, 5 ) <£. Frons and tuft of antennae at front Sanford-brown (ii) ; hairs of palpi orange-rufous (ii); a line of cream coloured (xvi) hairs between the antennae and posterior part of tuft at base of antennae cream colour; vertex with deep olive-grey (li) hairs; thorax above, ground colour of fore wing at basal half and at costal region on apical half mars-yellow (iii) ; hairs on tibiae and femurs, underside of thorax, upper and under side of abdomen and terminal area of fore wing ochraceous-salmon (xv); two flat crests at base of abdomen dark vinaceous-drab (xiv), first crest with a tuft of cream coloured hairs in centre; hairs at side of last abdominal segments cream coloured; anal tuft of spathu- lated hairs dark, vinaceous-drab mixed with black ; shaft of antennae covered with white scales and with teeth on inner and outer side for frd of shaft; rather long fasciculated, forwardly curved whitish hairs on each tooth. Fore wing; a broad white fascia along costa from base to near apex filling space between costa and vein 12 and extended near apex to vein 10; sub-basal line indicated by some dark scaling near Costa and below lower median; antemedial line faint, dark, most distinct below discal fold, angled inwardly at 1 b; a rounded blackish, macula, somewhat angled towards discocellulars, inner side with blackish scales, edge ill-defined by a blackish ring; orbicular large, angled inwardly on inner side, curved on outer side, coloured and defined 1 Father O’Neil informs me that the caterpillar of this species feeds on Brachy- stegia randi. Annals of the Transvaal Museum 161 as orbicular; postmedial line almost parallel to termen, cream coloured, curved at 8, nearly straight and very oblique to vein 4, then curved inwaidly to vein 2, then curved outwardly to 1 a, then oblique to inner margin; just beyond it and near tornus a slight hind-marginal tuft of dark vinaceous-drab hairs and a broader similar tuft just before it; some da,rk and white scales beyond post- medial line at apex above vein 8 ; a sub-terminal series of black lunules between the veins from vein 1 b to 8 ; some black scaling beyond each lunule ; cilia chamois (xxx), tipped with white from vein 3 to apex and getting mixed with dark scales towards tornus. Hind wing: ground colour cartridge -buff (xxx), slightly tinged with pinkish- buff (xxix) on the veins and on inner marginal area; cilia cream-buff. Under side: ground colour of both wings cream-buff; the costal fascia of fore wing cartridge-buff; fore wing tinged with vinaceous-buff (xl), thinly at inner marginal area; hind wing with the costa tinged with vinaceous-buff; cilia somewhat darker than on upper side. $. Ground colour of fore wing and thorax more vivid ochraceous-orange ; all markings less dark, ferruginous (xiv) ; a costal ferruginous fascia, less broad than the corresponding white fascia of antemedial line more distinct, ferruginous, double and filled in with some whiter scales, incurved at discal fold and below 1 b ; postmedial line less distinct, more curved between 2 and 5 and only clearly visible as far as vein 2; beyond it a white triangular macula at costa; sub-terminal lunules as in. <$; hind wing cartridge-buff, tinged with pinkish-cinnamon (xxix) ; under side of both wings deeply tinged with pinkish- cinnamon. Antennae simple; thoracic crest less pronounced, almost absent. $. var. Generally much darker, but lines and dark maculae obsolescent; thorax and colour of whole fore wing carob-brown (xiv) ; postmedial line very faint; white costal macula somewhat smaller; abdomen on upper and under sides cinnamon-rufous (xiv) ; hind wing on upper side and both wings on under side mikado-brown (xxix). This species comes close to niveiplaga Hmpsn. and argenteomaculata Aur. ; from the former it differs mainly in the orbicular and reniform being more rounded instead of elliptical, the macula below orbicular being large and not a mere blackish spot, the different postmedial line and the white macula being triangular, not wedge-shaped. From the latter it differs mainly in the thoracic tuft, which is quite long and produced even in the $ of argenteomaculata while the tuft of o’neili is short, triangular in the $ and almost absent in the $. Unfortunately of these two species only $s are known to me from descrip- tion and it will be interesting to know if their <^s will also have the broad white costal fascia. Exp. 46 mm.; $, 56 mm. Hab. Salisbury, i. 5. T8; 19. 5. '18; var. 24. 2. ’18, all bred by Father J. O’Neil, after whom I have much pleasure in naming this species. All in coll. Janse, kindly presented by Father O’Neil. Genus ICHTHYURA. (PI. I, fig. 6; PI. II, figs. 6-10.) Ichthyura Hiibn. Verz. bek. Schmett. p. 162 (1822). Hmpsn. Moths of India, 1. p. 172 (1892). Packard. Memoir of the Nat. Acad, of Sciences, vn. p. 123 (1895). Type anastomosis. Description from roseotincta. A, ?. Proboscis very short; head bent downwards; frons rounded and with a tuft of hair; palpi obliquely upturned; first joint short and curved; second 162 Annals of the Transvaal Museum joint somewhat curved on both ends, long, about three times first joints cylindrical; third joint short, somewhat pointed; all joints with long hair in front, with scales and hairs at the sides; eyes moderate, oval and covered with hairs; antennae rather short, less than half of costa; shaft curved, bipectinate in both sexes; pectination four times shaft in <$, two times in , on under side forming a straight line and getting very short beyond frd of shaft; legs rather short, especially the fore legs; fore tibia of $ with a pointed process on inner side, as long as tibia and covered with long dense hairs on outer side; in $ the process is narrow and the hairs are shorter; tarsi short and with dense tufts of long hairs on outer side; mid tibia with terminal spurs, longest inner spur about frd of whole tibia, outer spur somewhat shorter; hind tibia with four spurs, outer spur shortest; tarsi of last pair of legs covered with moderate hairs; thorax with a distinct crest and some long hairs at base of abdomen. Fore wing rather short; costa arched near base, then straight and curved towards apex ; apex obtuse ; outer margin erect, straight, curved towards tornus which is well rounded; inner margin curved at half; i b well forked at base; 2 from beyond |th lower median ; 3 from frd 2 to 4 ; 4 from lower angle and much curved; 5 from frd discocellulars, which are erect and somewhat curved; 6 slightly stalked with stalk of 7, 8, 9, 10 which comes from upper angle; 7 from stalk at frd of 8; 8 to apex; 9 from 8 beyond frd of 8; 10 from stalk, 8, 9 at before middle of 8; 11 from upper median at fth; 12 parallel to costa except near base where it is curved. Hind wing sub-triangular ; costa well arched; apex rounded; outer margin oblique, well rounded at vein 3, somewhat incurved between 4 and 6, 2 and 3; tornus and inner margin rounded; 1 a and 1 b slightly curved ; 2 from lower median at frd ; 3 from fth 2 to 4 ; 4 from lower angle; 5 absent, only a mere trace of it can be seen; discocellulars very oblique at lower half, then more erect; 6 and 7 stalked for about frd of 6; 7 well curved at tip; 8 slightly curved at base, then approximated to upper median beyond middle, but not touching it, then upcurved beyond frd and downwards towards tip. Range: N. America, Europe, Asia, Africa. The three South African species may be distinguished as follows : 1 a. Transverse lines almost absent; where the medial and postmedial lines should be the ground colour is slightly lighter ; fore wing vinaceous- lilac, irrorated with morocco-red, especially at the costal area violacearia b . Transverse lines distinct ; fore wing along the costa of ground colour or lighter ........... 2 2 a. Tuft of thorax in front and hairs of frons mummy-brown; hind wing as dark as fore wing; apical part of fore wing buckthorn-brown; reniform oval; sub-terminal line consisting of rather irregularly placed spots lentisignata b. Colour of thoracic tuft orange-cinnamon; hind wing lighter than fore wing; fore wing with the apical and terminal part lighter than ground colour; reniform round; sub-terminal line almost con- tinuous roseotincta ICHTHYURA LENTISIGNATA. Ichthyura lentisignata Hmpsn. A.M.N.H. 8. v. p. 495 (1910). I have one $ in my collection which agrees well with the description except for the oblique dark shade in the fore wing from postmedial line at Annals of the Transvaal Museum 163 discal fold to termen at vein 3. I have seen another specimen in Mr E. E. Platt’s collection, collected at Durban, in which the sub-terminal spots and the terminal black striae are entirely absent. Hab. Natal. Larva feeds on Tvimeria alnifolia Planch (E. E. Platt) ; pupates in silken cocoon between leaves of food plant. ICHTHYURA ROSEOTINCTA Spec. nOV. (PI. XIII, fig. 4; PI. I, fig. 6; PL II, figs. 6-10.) <$, $. Ground colour of thorax and fore wing congo-pink (xxviii), of abdomen on both sides, thorax, wings on under side, hairs on hind wing on upper side and legs pale vinaceous-fawn (xl) ; tuft of thorax in front, palpi at sides and costal area of fore wing densely irrorated with orange-cinnamon (xxix) ; some blackish hairs mixed with vinaceous-fawn hairs on frons; antennae with shaft pale congo-pink and a few orange-cinnamon scales, branches Rood’s brown (xviii). Fore wing, especially basal frd, irrorated with orange-cinnamon, terminal part only slightly, so as to have a light vinaceous-fawn (xl) tinge in certain light; sub-basal line outwardly oblique, straight from costa to lower median, then along lower median outwardly, then somewhat incurved and continued to inner margin ; antemedial line from costa to inner margin nearly parallel to sub-basal, excurved between costa and 1 b, then straight and oblique to inner margin; medial line from near costa and just touching the reniform, which is large, round and brownish-drab (xlv), in $ less distinct; medial line then well incurved as far as sub-median fold, then straight and outwardly oblique to join postmedial line at 1 b\ all these lines, a ring around the reniform and the postmedial line of ground colour; sub-basal line on inner side, antemedial line on outer side, and postmedial line on both sides rather darker edged with orange-cinnamon irroration; postmedial line nearly straight and inwardly oblique, slightly excurved at vein 6, gently incurved between 6 and sub-median fold, then somewhat excurved above 1 b and joining median line ; a sub- terminal slightly irregular, faint line, consisting of blackish scales and preceded by a line of ground colour, the blackish line somewhat excurved between 4 and beyond 3 ; cilia of ground colour, in £ tipped with black scales, in $ blackish at base. Hind wing with costal area, terminal half, inner marginal area and lower median thinly irrorated with orange-cinnamon ; cilia whitish with a few orange- cinnamon scales. Under side: fore wing pale vinaceous-fawn; costa narrowly and termen broadly tinged with sayal -brown (xxix) ; cilia as on upper side ; hind wing uniform pale vinaceous-fawn. Exp. 32 mm.; 34 mm. Hab. $ and $ type from Salisbury 8—10. 2. ’17, bred by Mr R. Jack, who kindly presented three specimens to me. ICHTHYURA VlOLACEARIA Spec. nOV. (PL XIII, figs. 5, 6.) cJ. Head, palpi, thorax, hairs on legs and ground colour of fore wing vinaceous-lilac (xliv), more or less densely, but finely, irrorated with morocco- red (i) ; fore wing with the transverse lines of ground colour and only faintly visible on account of morocco-red irroration; this irroration is very dense before postmedial line between the costa and vein 1 b; sub-terminal line very 164 Annals of the Transvaal Museum indistinct, oblique; antemedial line indistinct, preceded by a faint light line; medial line indistinct, morocco-red, straight from costa along discocellulars, then curved inwardly beyond origin of vein 2, then obliquely curved to post- medial line at inner margin; a faint dark round reniform beyond discocellulars; postmedial line slightly undulating, inwardly oblique, most distinct of all lines and preceded by a faint darker line of morocco-red ; beyond this line the morocco- red irroration has practically ceased so that this area has a beautiful, shiny vinaceous-lilac tinge, which is more or less continued between 1 b and inner margin; sub-terminal line, from costa to tornus, consists of some irregular morocco-red scaling between the veins; similarly the terminal line is indicated, but much more faintly; cilia vinaceous-lilac, gradually getting darker towards the tips; hind wing pure white, with the basal half hardly and the terminal half more densely irrorated with sorghum-brown (xxxix) ; cilia white. Under side of both wings white; fore wing and abdomen tinged with vinaceous -tawny (xxviii); cilia as above; hind wing slightly irrorated with vinaceous-tawny along costal area, abdomen above fawn coloured (xl). Fore wing as in $ along inner margin and beyond sub-terminal line, but beyond medial line from below lower median a triangular patch of lobelia- violet (xxxvii) densely irrorated with morocco-red before and less densely beyond postmedial line, which passes through it; upper edge of this triangle incurved and on terminal side connected with a similarly coloured shading before sub-terminal line, this patch extends from vein 4 to vein 8 ; the medial corner of this triangle is connected to the rounded and more distinct reniform by an oblique fascia; area, as far as sub-terminal line, not occupied by the lobelia-violet colour, with the ground colour ochraceous-buff (xv) (as also indicated in one of the <£s) and densely irrorated with morocco-red; hind wing of cream colour (xvi), terminal half ochraceous-buff and thickly irrorated with morocco-red; apical part of costa coloured like cilia of fore wing; cilia cream coloured with some dark scaling here and there. var. Ground colour of fore wing, head, thorax, abdomen and legs avellaneous (xl), somewhat violaceous on inner and terminal area of fore wing as indicated by the general markings of the typical $ ; here and there some lighter and darker irroration on the thorax and the fore wing ; transverse lines and orbicular absent; cilia of both wings of same colour as the wings. In this species the venation of the hind wing is slightly different from what it is in the other species, 3 is from frd 2 to 4, the stalk of 6, 7 is somewhat shorter, about ^th of 7 ; the typical $ and one also have four slight crests on the abdomen which are somewhat hidden in the long hairs in the $ ; the has also a forked anal tuft of hairs. Exp. 33 mm. ; ? , 42 mm. ; $ var., 40 mm. Hab. <$ type, Pinetown (Natal), 23. 6. T7; $ cotype, Malvern (Natal), 30. 3: T7; $ type, Krantzkloof (Natal), 12. 6. T6 (all bred by Mr E. E. Platt). $ var., Pretoria, 14. 7. '06 (bred, Janse). Larva feeds on Protea multibracteata Phillips; pupates in a loose silken cocoon between the leaves of the food plant (E. E. Platt). Genus PECTINOPHORA nov. (PI. I, fig- 7; PI. II, figs- H-17O Type noctuijormis. g. Proboscis well developed; palpi upturned, reaching vertex of head; first joint much curved and with long hairs in front; second joint about two times first joint, nearly straight, covered with hairs and scales, which are Annals of the Transvaal Museum 165 rather long in front; third joint short, about half first joint, somewhat oblique; eyes large, rounded, naked; frons oblique and with a tuft of moderate stiff hairs; thorax without crests and covered with hairs and hairlike scales; abdomen with lateral tufts of hairs at each segment; antennae a little over half of costa, bipectinated till tip, branches about six times shaft, somewhat curved forwardly at the tips and ciliated on both sides; first joint of shaft with a tuft of hairs; fore tibia with a twisted process nearly reaching end of tibia and almost hidden in the long hairs; femurs of all legs and of fore legs especially covered with long hairs on inner side, with scales at the sides; two spurs on the mid tibia, four spurs on hind tibia; spurs about ^rd of tibia in length, outer spurs |rd shorter than inner and all covered with hairs and hair- like scales and ending in a somewhat curved smooth point; tarsi with spines on inner side. Fore wing broad, sub-triangular; costa straight; apex somewhat rounded; termen nearly erect, curved and somewhat sinuate; tornus rounded; inner margin with a round lobe at basal half on which are two triangular tufts of scales and a small tuft just beyond postmedial line; 1 b simple at base, curved ; 2 from f rd lower median, curved ; 3 from |th 2 to 4 and curved ; 4 from lower angle, curved; 5 from just above middle of discocellulars; discocellulars out- wardly oblique, nearly straight; 6 from below upper angle; areole long, rather narrow; 7 from end of areole; stalk of 8, 9, 10 from end of areole; 9 from just beyond half distance end of cell to apex; 10 from |rd of stalk; 11 from about frd upper median; 12 parallel to costa. Hind wing triangular; costa well arched; apex rounded; outer margin very oblique and well rounded, slightly sinuate ; tornus rounded ; inner margin nearly straight except at 1 a where it is slightly lobed; 1 a straight, moderate; 1 b somewhat curved; 2 from well beyond half of lower median ; 3 from f th 2 to 4 ; 4 from lower angle ; 5 rather weak and from above middle of discocellulars ; discocellulars outwardly oblique; 6 and 7 stalked for Jth of whole of 6 and from upper angle; upper median well curved ; 8 curved at base, then curved towards upper median at frd and running parallel with it till frd, beyond this it is connected with the upper median by a rather long oblique bar, then curved towards apex. This genus is near to Hyperaeschra from which the fore wing differs in the more erect outer margin, shorter wing-length, the two tufts on the inner margin and veins 3 and 4 of both wings being farther apart and 6 of fore wing being remote from the areole. From Lophopteryx it differs in the upturned palpi, position of vein 6 of fore wing, and less sinuate outer margin. Pectinophora Noctuiformis spec. nov. (PI. XIII, fig. 7; PI. I, fig. 7; PI. II, figs, n-17.) c?. Hairs on head, thorax and fore legs, fore wing as far as postmedial line and along the whole costa shiny fuscous-black (xlvi); mid and hind legs fuscous-black; tarsi ringed with ivory-yellow (xxx) ; abdomen on upper and under side hair -brown (xlvi) ; fore wing with a faint basal black line ; medial line single from costa to lower median, then double and black, double lines filled in with russet (xv) and preceded by some french -grey (lii) scales; the medial line is oblique from costa and rounded along discocellulars, where there is some black scaling, then oblique to inner margin at tuft of scales, forming a tooth inwardly just below lower median and at 1 b; both lines fuscous-black, preceded by some french-grey scales ; two fine black lines in cell from near base to discocellulars, lower line shortest; a fine black line on terminal half of lower median and another below it from base to russet scaling and where the inner 12 i66 Annals of the Transvaal Museum median line is very faint; most veins indicated by black between medial and postmedial lines; postmedial line consisting of a series of four lines, from costa - oblique to 7 where they are curved inwards to vein 5, somewhat excurved on vein 3 and then oblique to inner margin ; first line black, second warm buff (xv) , third fuscous from costa to vein 6, then indistinct to vein 4, then russet, fourth line french-grey, represented by some scales only from costa to vein 4 ; veins beyond postmedial line irregularly scaled with black and french-grey; costal area beyond postmedial line and above vein 7 fuscous-black; area between 7 and 2 russet, with a warm buff line in the middle which is dentated inwardly on the veins and followed by some warm buff sub-terminal lunules between veins 2 to 7, and edged terminally by some black scales; between 1 b and 2, and 7 to 9 these lunules are only represented by black scales; area below vein 2 french- grey; a very narrow sinuate black terminal line with french -grey points on end of veins; cilia fuscous-black. Hind wing cartridge -buff, terminal half thickly irrorated with fuscous; inner marginal area with long cream -buff (xxx) hairs; cilia fuscous, tipped with cartridge-buff scales. Under side cartridge-buff; fore wing very densely irrorated and suffused with fuscous ; hind wing with some fuscous irroration along costa ; cilia as above but lighter; shaft of antennae fuscous-black, branches russet; abdomen at base with some fuscous-black long hairs in middle and cartridge -buff hairs at sides. Exp. $ type, 36 mm.; 2 <$ cotypes and 9 other $s. Hab. All the specimens were collected at Umtali (S. Rhodesia), from 5 — 15. 1. * 1 8. I have seen no other specimens of this species in other collections. In general appearance this species resembles very much the figure of Lophopteryx saturata given in Hampson’s Moths of India, vol. 1. fig. 102, except that the hind wing of that species shows strigae at the tornus of the hind wing and of these not a trace is seen in noctuiformis. Genus LOPHOPTERYX. (PI. I, fig. 8; PI. II, figs. 18-23.) Lophopteryx Steph. III. Brit. Ent. Haust. n. p. 26 (1892). Type camelina. Hmpsn. Moths of India, 1. p. 166. Spuler. Schmett. Eur. 1. p. 98. Schaus. Trans. Ent. Soc. Lond. p. 328 (1901). Pack. Mem. of the Nat. Acad, of Sciences, vii. p. 154 (1895). Description made from L. uniformis. $. Proboscis present but rather short; palpi short, porrect, just reaching frons, covered with scales, hairs and hair-like scales; first joint short; second joint about half length of palpus, slightly curved; third joint roundly pointed, nearly half of second joint; frons rounded; eyes naked, round, about as large as length of palpi; antennae with a small tuft at base; shaft short, less than half of costa, very much curved, bipectinated till tip; branches about three times shaft, bluntly pointed and covered with rather long cilia which stand rather far apart; fore tibia with a process reaching till end of tibia; this process is somewhat cylindrical and hollow at base, then, at befoie half its length it flattens and broadens so as to become leaf-like, then curved, with the apex roundly pointed, the outer side of the process is covered with scales and hairs and the edges of the leaf-like part have bristle-like hairs, the hollow side is towards the tibia; mid tibia with two spurs, hind tibia with median spurs as well ; the longer inner spurs are about ^rd of tibia, the outer spur is about Jrd Annals of the Transvaal Museum 167 shorter, they end in an acute inwardly curved point and the remainder length is covered with scale-like hairs; femora of all legs, especially those of hind legs,: rather thinly clothed with long hairs ; tibiae more thickly covered on outer side tarsi of mid and hind legs with some bristles on inner side and covered with hairs and scales. Fore wing broad ; costa straight ; curved at and beyond vein 1 1 ; apex rounded ; termen erect, crenulate ; tornus sharply rounded ; inner margin with a large rounded lobe at basal half on which is a large tuft of hairs; another but smaller tuft of hairs beyond postmedial line; 1 b somewhat curved and in- distinctly forked at base; lower median curved upwardly; 2 from fth lower median ; 3 from f rd 2 to 4 ; 4 from lower angle and well curved ; 5 from middle of discocellulars and somewhat weak ; discocellulars erect, curved below and above vein 5; 6 from areole at f-th of areole; areole very long, about half length of vein 10, narrow; 7 from end of areole; 8, 9, 10 stalked; 9 from 8 at |rd whole length of 8; vein 10 from stalk of 8, 9 at little beyond areole; 11 from upper median at fth, upper median curved upwards; 12 parallel to costa. Hind wing sub-triangular; costa curved; apex well rounded; termen oblique, well rounded, crenulate; tornus rounded; inner margin straight; 1 a and 1 b slightly curved ; 2 from lower median at fth ; 3 from fth 2-4 ; 4 from lower angle, well curved; 5 rather weak and from above middle of disco- cellulars; upper discocellular erect, lower oblique and both somewhat curved; an indistinct veinlet into the cell beyond vein 5 ; 6 and 7 from upper angle, stalked for about length of upper discocellular; upper median well curved; 8 gently curved at base, then downwards to approach and run parallel with terminal half of upper median, then curved upwards to apex; at before half of upper median there is an indication of a bar on this vein and vein 8. I have little doubt that my specimens have to come in this genus, though in this species the antennae are well pectinated, even in the $, while in the other species of Lophopteryx they are serrate in the all other characters agree so well, that I keep the South African species here. The figure of the wing venation given by Packard differs in a few characters from my specimens; it has the areole much shorter, 6 shortly stalked in the fore wing and no veinlet in the cell, and the inner margin has a pointed lobe ; in the hind wing the outer and inner margins are not as straight and the stalk of 6, 7 is shorter in my specimens. The illustration of the fore leg differs from mine in the tarsi and the process being much shorter in the American species, but it is not possible to recognise the shape and structure of this process even in the denuded specimen. Lophopteryx Uniformis. (PI I, fig. 8; Pl. II, figs. 18-23.) Lophopteryx uniformis Swinh. A.M.N.H. 7. xix. p. 208 (1907). The description given by Swinhoe is rather short but fits my specimens in every respect, except that the dorsal side of the abdomen is not whitish tinged with pink in my specimens, but more yellow. There are, however, several details that can be seen which have been omitted. I therefore give a fairly full re-description of my specimens. $. Head, thorax, fore wing, tibiae and femora of all legs bone-brown (xl), the thorax and the fore wing here and there tinged with chestnut (ii) ; shaft of antennae bone-brown, pecten clay colour (xxix); abdomen above cinnamon- buff (xxix) slightly tinged with hair-brown (xlvi), on under side nearly entirely hair-brown; two white spots on meso-thorax. Fore wing with a white spot at base above 1 b and a few white hairs below it; antemedial line blackish, pre- 12 — 2 i68 Annals of the Transvaal Museum ceded by antimony-yellow (xv) faint lunules and some scales of same colour, beyond it above lower median and as far as discocellulars, beginning at |rd of costa, then well curved outwardly and then obliquely inwardly to inner margin, where it ends in a few antimony -yellow scales at base of first scale tooth; postmedial line blackish and followed by irregular antimony -yellow lunules and with some similar scaling before it between 2-4, and 5-6 and again beyond it between veins 3—4; the line begins at frd of costa, runs then nearly parallel to costa for half the distance to apex, then oblique and nearly straight to base of second scale-tooth, very much dentated outwardly at veins 2-8; sub-terminal line represented by a series of black spots between veins 3-8, preceded by some yellow scaling; cilia of ground colour, but with a lighter base. Hind wing cartridge-buff (xxx), exterior half gradually shaded with hair-brown, most intensely at outer margin; cilia cartridge-buff, tinged with hair-brown. Under side: tarsi cartridge-buff ringed with bone-brown; fore wing uniform hair-brown; cilia hair-brown at base, then bone-brown and with a few cartridge- buff scales between the veins indicating a semilunar line ; hind wing as on upper side. Exp. 62 mm. Hab. Umtali (S. Rhodesia), 9. 1. ’18 (Janse). Genus PSEUDORETHONA. (PI. I, fig. 9; PI. Ill, figs. 1-6.) Type albicans Wlk. Pseudorethona gen. nov. $. Proboscis very rudimentary; palpi short, scarcely as long as width of eye, porrect; first and second joints of equal length; third joint frd of second, globular, covered with long, spreading hairs; eyes large, rounded, glabrous; antennae bipectinate as in Centra but branches shorter, about four times shaft, gradually getting shorter towards tip; basal part of shaft as in Cerura\ fore tibia with a process as in Centra but slightly shorter and more straight; mid tibia with two terminal pointed spurs of about fth of tibia in length; hind tibia with two pairs of pointed spurs about frd of tibia in length; spurs rather long and thin, with two very short rows of short teeth near apex only; all legs covered with long, spreading hairs; tarsi with hairs and scales; mid and hind tibiae with small scale tufts at middle and terminus. Fore wing rather long and narrow; costa evenly incurved; apex rounded; termen oblique, evenly curved; tornus well rounded; inner margin somewhat lobed near spur of 1 b\ 1 b with a fork at base and a spur which, however, does not reach inner margin; 2 from |th lower median; 3 from fth distance 2 to 4; 4 from lower angle; 5 rather weak, from above middle of discocellulars and with an indication of a continued veinlet in the cell; 6 from end of areole, which is nearly 4 times the breadth; 7 from frd stalk of 8, 9; 8 and 9 on a stalk of frd 8, originating from end of areole; 10 from just before end of areole; 11 from frd upper median, somewhat curved inwardly near areole; 12 parallel to costa. Hind wing sub-triangular; costa nearly straight; apex rounded; termen well rounded; tornus with a rounded lobe at 1 b; inner margin evenly curved; 1 a and 1 b nearly straight ; 2 from lower median at frd ; 3 and 4 from lower angle ; 5 faint, but distinct, from well above middle of discocellulars; discocellulars oblique and faint; 6 and 7 on a stalk of nearly half of 6; 8 a little upcurved till frd upper median, where it has a little bulge to suggest rudiments of a bar, then approximated and parallel to near end of cell, then upcurved towards costa. Annals of the Transvaal Museum 169 The type of this genus has been placed by Walker in Rethona, but Sir George Hampson kindly informed me that this genus, of which strigosa is the type, has to come in the Limacodidae. R . albicans, Wlk. was placed by Prof. Aurivillius in the genus Cerura when he identified some of my specimens, but the presence of four spurs on the hind tibiae is sufficient to remove this species from Cerura. I have therefore been obliged to create another genus for it. There are, however, many points of resemblance between this genus and Cerura ; the venation on the whole, the process of the fore tibia, the shape of the spurs, the pectination of the antennae and the type of marking on the fore wing all point to close relationship. PSEUDORETHONA ALBICANS. (PI. I, fig. 9; PI. Ill, figs. 1-6.) Rethona albicans Wlk. Cat. v. p. 1043 (1855). The description given by Walker is very short and much too vague, so I think it useful to re-describe the species. <$. Whole insect pure white; hairs on frons, on palpi and at sides of head fuscous (xlvi); thorax, and fore wing at basal, medial area, and postmedial area above vein 6, light patch at apex and tornus tinged with pale dull green- yellow (xvii) ; lines on head, thorax and wings black ; a few black scales on frons and before antennae ; patagia and tegulae edged with black ; a crest beyond each tegula and tipped with black; a central crest on metathorax also tipped with black; abdomen above hair-brown (xlvi). Fore wing with a broad basal line from costa till 1 b, dentate on the veins ; a double sub-basal line from costa to inner margin, curved and much angled at the veins, space between them pure white; antemedial line double, more erect, very much angled and dentated on the veins, and space between them pure white, lines farther apart near costa than at inner margin ; space between outer sub-basal line and inner antemedial line, densely irrorated with black except a round spot in the cell which is tinged with pale dull yellow-green; medial line double, broad near costa, as is the case in the outer antemedial line, then narrow ; inner line parallel to outer antemedial as far as vein 2, then touching antemedial and forming between them a round spot tinged with the light greenish, a black irroration between the lines at inner margin ; outer medial line curved outwards below costa touching outer side of orbicular, which consists of two black strigae filled in with the pale greenish, then almost straight to vein 2, then dentate inwardly on plical fold and 1 b and touching antemedial at inner margin; postmedial beginning as a broad oblique streak from before apex, which is sharply defined on the outer side, then a little narrower and less defined to near medial line at vein 4, then outwardly oblique to vein 2, then curved inwards to plical fold, then curved outwards above 1 b, then erect to inner margin; a black irroration beyond it reaching the termen between veins 7 and 2 except for a triangular patch from vein 7 to 6 which remains tinged with the pale green; a black irroration at tornus below plical fold; a black fine terminal line, rather thicker between the veins; cilia white, tinged with pale green and checkered with black between the veins. Hind wing pure white, except some fuscous black scaling at the tornus; a fine fuscous-black terminal line interrupted by the white at the veins; cilia white. Under side: wings, thorax and abdomen and hairs on legs pure white; tarsi black, ringed with white; fore wing with costa broadly edged with black and with some white irregular marks at basal, antemedial, medial, and postmedial areas ; terminal half of wing from costa to vein 2 irrorated with fuscous, only 170 Annals of the Transvaal Museum leaving an apical elongate white patch; both wings with a terminal black line; cilia of both wings as above; hind wing with a diffused fuscous costal shading at antemedial and medial areas; a well-defined costal postmedial fuscous patch; a faint fuscous striga on discocellulars and a very faint fuscous postmedial line. $. Like but postmedial line much more incurved and more densely irrorated with fuscous, hind wings tinged with cinnamon -brown (xv) ; all lines less defined ; under side, the whole fore wing is suffused with cinnamon-brown except some terminal and apical white patches; hind wing with terminal half suffused with cinnamon -brown except near termen; and the postmedial line which is left white ; basal half irrorated with cinnamon-brown ; striga on disco- cellulars and terminal line cinnamon-brown. Exp. . Antheua Peringueyi spec, nov.1 <$, $. Head and thorax with fuscous (xlvi) and white hairs almost equally mixed, so as to produce a dark grey colour when not viewed with a magnifying glass and of the same tint as the general colour of the forewings; abdomen avellaneous (xl) ; thorax in $ with a small tuft of scales on mesothorax ; fore wing whitish, rather densely irrorated with fuscous, especially at base and along costa; indications of a dentate fuscous medial line; a fuscous striga at end of cell; an outwardly dentate postmedial line, indistinct from costa to vein 7, then more distinct till vein 2, then indistinct; indications of a sub-terminal diffused irregular line ; a distinct zigzag terminal line ; cilia have general colour of fore wing. Hind wing light buff, distal part densely irrorated with fuscous and a medial line indicated by a fuscous irroration Under side: both wings whitish; fore wing with a rather dense fuscous shading between postmedial and sub-terminal lines, broad near costa where it has two small, white, oblique, strigae and gradually getting more narrow till it ceases beyond vein 2; hind wing with medial line indicated by a fuscous irroration; some fuscous irroration beyond it along costa till near tornus. Antennae bipectinated in $ till the tip; in $ simple; branches antimony- yellow (xv). Exp. 3, 40 mm. ; $ , 46 mm. Hab. Capetown, March 1917 (P. C. Keytel). type in S. African museum; $ cotype in coll. Janse. Antheua Mixta spec. nov. (PI. XIII, fig. 9.) Frons of head and thorax on under side apricot-yellow (iv) ; thorax above, hairs on palpi and legs on outer side marguerite-yellow (xxx), on palpi mixed with black hairs ; tibiae on inner side covered with black hairs mixed with white ; tarsi black ringed with white ; a stripe on each .side of thorax on upper side and hairs covering first segment of abdomen buff-yellow (iv), abdomen on upper side (except the last two segments, which are ringed black and buff- yellow) ochraceous-orange (xv); a lateral black line, checkered with whitish on abdomen ; abdomen white on under side ; branches of antennae ochraceous- orange. Fore wing white and with broad fasciae, consisting of black and fuscous (xlvi) scales which are thickly covered with marguerite-yellow hair-like scales, so as to give a greyish tinge to the fasciae; costal fascia from base to apex, as broad as from costa to a little below upper median and having a whitish fascia between veins 7 and 8; second dark fascia from discocellulars to outer margin and filling the whole space from costal fascia between veins 5 to 7; a short, triangular terminal fascia of ground colour between veins 5 and 6; third fascia from base of wing to outer margin, joining the costal fascia till well beyond base and forming a straight edge in cell above the lower median, then along vein 4 for half that vein, then above vein 4 to outer margin ; lower edge is well below lower median and vein 2 and parallel to discal 1 This species was unfortunately not on hand at the time when the coloured figures were made, so that it was impossible to figure this species. Annals of the Transvaal Museum 183 fold; two short, white, sub -triangular outer marginal fascia between 2 and 3, 3 and 4; fourth inner marginal fascia from base to tornus, extending to near discai fold; inner margin with a whitish edging; cilia white, checkered with a bunch of black scales on each side of veins 1 b to 8, but remaining white just opposite the veins. Hind wing with buff -yellow hairs on inner margin; some fuscous scaling on veins 2 to 8 at terminal area; an indistinct black terminal line; cilia white and checkered as in fore wing. Under side of both wings white and with a terminal black line and some dark scaling on the veins at terminal area. „ $ . Of stouter build ; the antennae are unfortunately missing in my speci- men, but are probably simple as in the $ the branches are already very short, only a little oyer thickness of shaft. The areole of the fore wing is in this species normal. Exp. cL 48 mm. ; $ , 53 mm. Hab. <$ type from Charlestown (Natal) in coll. Janse; $ cotype from Likhoele (Basutoland), collected by Dieterlin, in coll. S. African Museum. Species omitted : A . spurcata Wlk. = Eutimia marpissa Wllgrn., which I place in the genus Zana. Genus DESMEOCRAERA. (PI. I, fig. 21; PI. IV, figs. 1-12.) Desmeocraera Wllgrn. Kongl. Vet. Akad. Handl. 2. Bd. v. No. 4, p. 52 (1865). Auriv. Ofv. Vet. Akad. Fork. No. 9, p. 1050 (1900). Type interpellatrix . Description from interpellatvix . <£, Proboscis almost absent; palpi upturned, reaching vertex of head; first joint short, about fth of second joint; second joint long, curved at base, gradually tapering; third joint very minute, in some species absent; all joints covered with rather short hairs and scales; eyes rounded, large, naked; frons rounded, covered with hairs and scales; antennae in $ about frd of costa in length, in $ only fth; basal joint globular and with a tuft of hairs and scales in front; bipectinated for fth, then serrate; pectination rather long, in middle about 5 times shaft; each branch with a terminal bristle and ciliated anteriorly; fore legs with a rather narrow, somewhat pointed process, reaching till end of tibia; mid tibia with two spurs as long as fth of tibia; spurs with a terminal serrate edge on inner side and covered with hairs; hind legs with four spurs, about frd of tibia and in structure as on twiddle legs; all femora covered with long hairs ; tibia of fore leg with hairs and scales, of other legs with hairs only; tarsi with dense, short, stout hairs. Fore wing sub-triangular, costa almost straight, well curved towards apex, outer margin somewhat rounded, oblique, inner margin straight, apex and tornus rounded ; 1 b forked ; 2 from just beyond frd lower median ; 3 from before lower angle; 4 from angle; 5 from middle of discocellulars, which are straight and erect; 6 from upper angle or more or less stalked with 7, 8, 9, 10; stalk from upper angle; 8 from just beyond frd of 7 ; 9 from before frd of 8 ; 10 from middle of origin of 7 and 8; 11 from beyond fth upper median; 12 straight, parallel to costa. Hind wing somewhat circular; costa, termen and inner margin well curved ; apex and tornus well rounded ; 1 a and 1 b straight ; 2 from frd lower median; 3 from just before lower angle; 4 from angle; 5 from above middle of discocellulars, which are oblique ; 6 and 7 on a stalk of f of 6 ; 8 free, first upcurved, then curved towards upper median and parallel to it till near upper angle, then curved towards costa. The species placed by me in this genus were placed by nearly all authors 13—2 184 Annals of the Transvaal Museum in the genus Stauropus. All species known to me differ, however, from Stauropus , in having four spurs on the hind legs, having vein 10 of fore wing from beyond vein 7, having vein 8 of hind wing not touching the upper median, as it does in Stauropus for over half of cell, and in having rather long and upturned palpi. I only keep mediata Wlk. in Stauropus. A $ specimen of Desmeocraera interpellate in my collection has been identified by Prof. Auri- villius and the same distinguished entomologist drew my attention to the fact that atriguttata and calliope have to be placed in Desmeocraera on account of the four spurs on the hind legs. As the type of D. interpellatrix Wllgrn. is probably in the Stockholm Museum and the species is apparently well known to Prof. Aurivillius, I have no doubt that his identification is correct, though I cannot confirm this for certain on strength of the descrip’tion. It is true, that my three specimens agree well with the description, but the description is of course far from complete. The only difference of importance is, that the ground colour of my specimens is not grey, but greyish green ; seeing, however, that this green colour is not easily preserved, it is quite well possible that Wallengren’s specimens were rather faded. Desmeocraera hierax Dist. has to be taken out of this genus and has to come in the genus Pararethona as the wing venation and other structural characters show little direct relationship to Desmeocraera and why Distant has placed it there is a mystery to me. Key to the South African species: 1 a. Ground colour of fore wing green or irrorated with green . . 2 b. Ground colour of fore wing white, irrorated with grey or with brown, or ground colour greyish-green ...... 9 2 a. Ground colour of fore wing malachite-green (xxxii), orbicular and reniform defined by narrow silvery white lines ; a fuscous sub-terminal patch between veins 5 and 6; some silvery white edging on sub- terminal, antemedial, medial and postmedial lines . calliope p. «i86 b. Fore wing not malachite-green, but lighter ...... 3 3 a. Sub-terminal line represented by a more or less irregular fine, often faint line ........... 5 b. Sub-terminal line represented by spots between the veins only . . 4 4 a. All markings on the fore wing somewhat diffused, sub-terminal spots most clearly defined of all; postmedial line very diffused; wing thickly irrorated with biscay-green (xvii) ; hind wing brownish atriguttata p. 186 b . Markings on the fore wings sharply defined and edged with white, post- medial line well defined, narrow, black and very irregular; ground colour of fore wing pale olivine (xxxii) ; hind wing whitish thalassina p. 187 5 a. Ground colour of fore wing water-green (xli), three diffused rounded spots, the smallest at lower angle, a larger one before this on lower median and one below lower median at before origin of vein 2 tripuncta p. 189 b. No diffused spots on fore wing as mentioned above; ground colour of fore wing water-green or slightly darker ...... 6 6 a. All markings, except sub-terminal line, very diffused, almost absent; antemedial and postmedial lines indicated by some dark scales, mixed with yellow scales; indications of white-ringed orbicular and reniform ....... vernalis p. 186 b. Markings more distinct; no yellow scales mixed with it 7 7 a. Postmedial area irrorated with dark green scales; hind wing of $ whitish, veins beyond the sub-terminal line not streaked with black ........ interpellatrix p. 185 Annals of the Transvaal Museum 185 b. Postmedial area not much darker than ground colour; hind wing of $ thickly irrorated with brown; black streaks beyond sub-terminal line as far as termen ......... 8 8 a. Hind wing with white streaks on the cilia; a distinct black irroration in middle of postmedial area, edged on the inner side by yellowish- green, on outer side by white, below vein 4 connected with a black patch to sub-terminal line and below vein 2 with a black streak towards base, which is interrupted by green at antemedial area platti p. 190 b. Cilia of hind wings not interrupted by white streaks; no distinct irroration beyond postmedial line, though in the $ there is an indication of it ; there is, however, no dark patch below vein 4 ; often there is a continuous or interrupted black streak, broad or narrow, below lower median from base to postmedial line; hind wing with postmedial line distinctly indicated near costa by two black streaks as far as vein 6 ....... varia p. 187 9 a. Sub-basal line obliquely outwards towards costa .... 10 b. Sub-basal line obliquely outwards towards inner margin . . . 11 c. No sub-basal line at all ....... incana p. 189 10 a. Ground colour of fore wing thickly irrorated with drab (xlvi) ; some yellow hairs at ante- and postmedial lines ; hind wing irrorated with brown ......... vernalis p. 186 b. Ground colour of fore wing light mouse-grey (li) ; hind wing pure white except at costa near apex .... canescens p. 188 11 a. Sub-basal area rather dark, well defined by a double black line; cilia greyish; pecten of antennae in over 3 times shaft basalis p. 186 b. No black lines defining sub-basal area, which is of ground colour, but defined by a single brown line; cilia of hind wing almost white; pectination of antennae in $ short, about two times shaft . . 12 12 a. Antemedial line nearly straight; postmedial line somewhat incurved at lower angle of cell, beyond which it just passes steniptera p. 191 b. Antemedial line angled outwards in cell, inwards at submedian fold; postmedial line curved outwards beyond vein 3, passing a good distance beyond discocellulars .... pergrisea p. 191 Desmeocraera Interpellatrix. (PI. I, fig. 21; PI. IV, figs. 1-3.) Stauropus interpellatrix Wllgrn. Wien. Ent. Monats. iv. p. 164 (i860). Desmeocraera interpellatrix Wllgrn. Kongl. Vet. Akad. Handl. 2. Bd. v. No. 4, p. 52 (1865). I have two <^s of this species in my collection, one from White River (Aug. 1909), and one from Krantzkloof (Aug. 1916); and one $ from Eshowe (Zulu- land) (Jan. 1916). One $ in Transvaal Museum from Durban (Febr. 1908). As remarked in my notes on the genus, my specimens are not grey, but tinged with olivine (xxxii) green; however, for reasons given above, I have little doubt as to the correctness of the identification, as moreover, the species is very distinct from any of the other species known to me. The third joint of the palpus of this species is very minute and only visible between the hairs at high magnification. i86 Annals of the Transvaal Museum Desmeocraera Atriguttata. (PI. IV, fig. 8.) *’ Stauropus atriguttata Hmpsn. A.M.N.H. 8. v. p. 465 (1910). I have <£s and $s in my collection from Durban caught in Jan. and May. Sir George Hampson only described the $ of this species. The £ is much smaller, 44—50 mm., but is otherwise identical with the $. Both sexes have a green patch of hairs and scales on the upper side of the last abdominal segments, a character not mentioned in the description. Larva feeds on Mimusops obovata Sond., Chrysophyllum viridifolium Wood, and C. natalense Sond, pupates in a brittle earthen cocoon (E. E. Platt). The third joint of the palpus is longer than in interpellatrix . Desmeocraera Calliope. (PI. IV, figs. 4, 5.) Stauropus calliope Hmpsn, A.M.N.H. 8. v. p. 466 (1910). Desmeocraera ianthina Aur. iv. Only $s in my collection, all from Durban in J an., March, J une, November. The spurs of the hind legs, especially the terminal pair, are shorter, more curved at the point and the toothed ridge is not visible; the third joint of the palpi also seems to be absent in this species. Larva feeds on Chrysophyllum viridifolium Wood (H. A. Green). Spins a cocoon in chinks of bark, some distance from the ground (E. E. Platt) . Desmeocraera Vernalis. (PI. IV, fig. 6.) Desmeocraera vernalis Dist. A.M.N.H. 6. xx. p. 205 (1897). Dist. Ins. Transv. iv. p. 89, PI. IV, fig. 1. My specimens agree well with the quite good figure and the rather poor description and I have seen a specimen identified by Distant himself. I find that the yellow scales mixed with black scales on the antemedial and post- medial lines, as mentioned in the key, are very reliable characters. In the grey variety these scales are also present though the lines themselves are very indistinct in that form. The general ground colour of the fore wings is not green, but greyish. Hab. In coll. Transvaal Museum from Durban in Sept, and Nov., and from Pretoria in Jan. Larva feeds on Combretum apiculatum Sond., pupates in a brittle earthen cocoon (E. E. Platt). In coll. Janse from Durban (March); Karkloof (Jan.); Waterval onder (Nov.); Barberton (Nov.); Eshowe (Zululand) (Jan.). Desmeocraera Basalis. Desmeocraera basalis Dist. A.M.N.H. 7. iv. p. 361 (1899). Dist. Ins. Transv. iv. p. 89, PI. IV, fig. 14. Three specimens, one in the Transvaal Museum and two in my own col- lection, agree well with the figure except that the sub-basal lines do not go deeply inwards at vein 1 b as shown in the illustration. One of my specimens Annals of the Transvaal Museum i87 (from Emangeni) shows the basal patch rather lighter than the ground colour of the fore wing and not darker as in other specimens and as shown in illustra- tion. The description given by Distant is much too vague and if no good figure was given and the species was not as strikingly marked as it is, I would not have been certain of my identification. In my key several useful characters are given, but the direction of the sub-basal line, being oblique outwardly towards inner margin and the area before it being filled in with almost black scales, separate this species at once from the grey form of vernalis for which it might be mistaken, especially as basalis has also yellow scales between the black scales. The costa of the fore wing is also less curved at the base and the termen is more oblique than in vernalis. Hab. In Transvaal Museum one $ from Nelspruit (Nov. 1917, Dr H. G. Breyer) ; in coll. Janse 3 <$s from S. Rhodesia. (Emangeni, Jan. 1918; Umvuma, Dec.) Desmeocraera Thalassina. (PI. IV, fig. 9.) Stauropus thalassina Hmpsn. A.M.N.H. 8. .v. p. 470 (1910). My two $ specimens agree in every respect with the description and both specimens come from the recorded locality (Salisbury, bred by Father O’Neil in March and May). Father O’Neil kindly sent me a drawing of the caterpillar from which the specimens were bred. This larva had 6 long front legs and four pairs of ab- dominal legs on the 3rd to 6th abdominal segments; the first and second abdominal segments had each a short hump and the last abdominal segments were turned up and the very last segment ended in two bristles as long as the breadth of that segment and placed in the shape of a V. It pupated for 20 days. The front tarsi of the moth show some brushes of spreading hairs which are not found in any other species of this genus, also the palpi are shorter and have spreading hairs. Desmeocraera Varia spec. nov. (Pi. XIII, figs, n, 12, 13; PI. IV, fig. 7.) A. Thorax, ground colour of fore wing and costal area of hind wing, water- green (xli) ; vertex of head, palpi in front, legs, meso- and metathorax and abdomen on the under side cartridge-buff (xxx) ; sides of palpi, hairs of pro- thorax on under side, on tibia and femurs of fore legs and rings on tarsi of all legs natal -brown (xl) ; abdomen on upper side except the terminal segments and the remainder of the hind wings on upper side verona-brown (xxix) ; terminal segments of abdomen water-green; at base of abdomen a tuft of water-green hairs and scales mixed with black scales and hairs on the sides; shaft of antennae with white and natal -brown scales ; branches clay colour (xxix) . Fore wing with all lines black, in some specimens beginning at costa with a natal-brown patch; sub-basal line confluent with basal line and erect till lower median, then excurved below lower median till antemedial line, some white and dark green scales beyond it; antemedial line double, very irregular and often obsolete, inner line heaviest, outer line thinner and preceded by a white line ; a whitish patch beyond it below lower median ; medial line absent from costa to vein 2, then faint and dentate on 1 b ; orbicular and reniform edged with white scales; postmedial line double, very irregular and much dentated on the veins, excurved at vein 7, incurved between 7 and 4, then erect to inner margin; space between the lines at veins 2 to 4 filled in with cinnamon-buff i88 Annals of the Transvaal Museum scales and whitish scales beyond it; three costal blackish patches beyond postmedial line and some black scales below these patches; sub-terminal line distinct, narrow, irregular and from ib to 8. incurved between 4 and 6, indented at 2, 3 and 1 b where it has some black scales at the tornus; some black strigae from termen to near sub-terminal line on veins 2 to 8 ; cilia consisting of black and brown scales, checkered with cartridge-buff scales at the veins. Hind wing with a postmedial double line of black scales from costa to vein 6, inner line well defined, outer one very diffused; cilia uniform of natal-brown and cartridge- buff scales. Under side : the whole fore wing, except inner marginal area, cartridge-buff, irrorated thickly with cinnamon-brown (xv) ; hind wing cartridge -buff, with some cinnamon-brown scales near apex ; cilia of fore wing with cinnamon -brown tips, cilia of hind wing of ground colour. In some $ specimens the dark irroration at postmedial area is very much as in $ and a well-defined outer whitish line is present beyond the black sub-terminal line. In others again the antemedial, medial and postmedial lines are very obsolete. $. Like but more dark irroration in fore wing beyond postmedial lines, and here and there more chromium-green (xxxii) scales; whitish scales also more pronounced and more raised; postmedial dark irroration edged on the outer side by a crenulate whitish line, inwardly dentate on the veins; black basal line below the lower median continued till beyond postmedial line. In hind wing the irroration is more dense and extended till the inner margin. On the under side the hind wing is as much irrorated as in the fore wing. One $ cotype (in Transvaal Museum) has the postmedial patches between veins 2 and 4 dark. Exp. $ type, 38 mm.; $ type, 50 mm. Hab. type from Durban, 21. 9. To (Leigh); $ type, Durban, 18. 9. T4 (bred by E. E. Platt) ; cotypes and other specimens from the same locality in April, July, Oct. and Nov. Probably very close to octoginta, but differs from it in the black streak of the median fold not going beyond postmedial line, in the presence of black streaks on the veins from sub-terminal line to outer margin, and a few other characters. Var.