Cyphomyrmex faunulus Wheeler, 1964

Kempf, W. W., 1964, A revision of the Neotropical fungus-growing ants of the genus Cyphomyrmex Mayr. Part I. Group of strigatus Mayr (Hym., Formicidae)., Studia Entomologica (N. S.) 7, pp. 1-44 : 23-25

6. Cyphomyrmex faunulus Wheeler , n. stat.

(Figs. 5, 15, 31, 47)

Cyphomyrmex bigibbosus faunulus Wheeler , 1925: 44-45 (Worker, female; British Guiana: Kartabo, Camaria; Bion.). - Weber, 1938: 187 (British Guiana: Forest Settlement, Mazaruni River). - Weber, 1946: 124-126, pls. 2. 8 (British Guiana; Bion.).

Cyphomyrmex bigibbosus : Weber, 1938 (nec Emery, 1894): 203 (Bolivia: Convendo, Huachi, Lower Rio Madidi, Cachuela Esperanza, Won, Riberalta). - Weber, 1940: 413 (Key). - Weber, 1945: 14-16 (British Guiana, Trinidad; Bion.). - Weber, 1946: 122-124, Pl. 1 (British Guiana: Oronoque River;.Bion.).

Cyphomyrmex bigibbosus petiolatus Weber , 1938: 187-188 (Worker, female; Bolivia: Lower Madidi River, Cavinas). - NOV. SYN.

Types. - Workers and female of faunulus (MCZ) from Kartabo (type locality) and Camaria, British Guiana, and a worker of petiolatus (NAW) from Lower Rio Madidi, Bolivia, examined.

Worker. - Total length 3.4-4.0 mm; head length 0.80- 0.98 mm; head width 0.75-0.85 mm; thorax length 1.01-1,20 mm; hind femur length 0.96-1.20 mm. Yellowish-brown to fuscous-ferruginous; opaque. Very close to bigibbosus with the following differences:' 1. Color more uniform, and averaging larger in size. 2. Occipital corners auriculate and projecting (Figs. 5, 47). 3. Thorax: midpronotal tubercle low; anterior mesonotal tubercles very strong and conical, posterior tubercles low and feeble; anteroinferior corner of pronotum less acute; mesoepinotal constriction deeper (Fig. 15). 4. Petiole slightly broader, anterior corners angular in dorsal view. Postpetiole deeper, the posterior border not excised mesially nor flanked by prominent tubercles (Fig. 31.).

Female. - Total length 4.2-4.8 mm; head length 0.96- 1.07 mm; head width 0.83-0.93 mm; thorax length 1.25-1.41 mm; hind femur length 1.07-1.28 mm. Resembling the worker with the differences of the caste. Quite close to auritus , with the following distinctive features: Carinae flanking frontal area obsolete, carinae of vertex extremely weak. Sides of head lacking a subcarinate ridge connecting low supraocular tubercle with inferior occipital corner. Pronotum with low and blunt lateral tubercle, midpronotal tubercle obsolete. Scutum with only shallowly impressed Mayrian furrows. Paraptera postero-laterally strongly and acutely dentate, teeth facing caudad. Scutellar teeth longer than their width at base. Epinotal teeth minute to nearly obsolete. Postpetiole not strikingly transverse, middorsal longitudinal impression shallow. Tergum I of gaster lacking lateral and mesial paired longitudinal ridges. Appressed hairs minute and highly inconspicuous.

Distribution. - This species ranges from Trinidad over the Guianas through the Amazon river valley to the Beni river valley in western Bolivia.

Specimens examined: British Guiana: Kartabo (W. M. Wheeler) 2 workers, 1 female (lectotype and paratypes of faunulus ) (MCZ); Camaria, 1 worker (paratype) (MCZ); Oronoque River (N. A. Weber) 1 worker, 1 female (NAW); Forest Settlement, Mazaruni River (N. A. Weber) 2 workers (WWK). - Brazil, Amazonas State: Manaus (K. Lenko) several nest series with many workers, several females and males (DZSP, WWK), Benjamim Constant (K. Lenko) 1 nest series with many workers and 1 female (DZSP, WWK). - Bolivia: Covendo (W. M. Mann) 1 worker (WWK); Ivon, Beni River (W. M. Mann) 3 workers (WWK); Lower Madidi River (W. M. Mann) 1 worker (syntype of petiolatus ) (NAW).

Discussion. - It was shown above that the typical bigibbosus is identical with the race later described by Weber as tumulus . It remains here to decide the fate of the other forms hitherto recognized in the tightly knit complex: " bigibbosus " (of authors, not of Emery), faunulus and petiolatus , neatly differentiated in Weber's (1940: 413) key. From the material which I was able to gather, I reached the conclusion that they are not separable, but constitute a single species, that takes the oldest available name viz. faunulus .

C. faunulus was thought to be a rather small race of what had wrongly been taken as the typical bigibbosus . The type series of the former is indeed on the lower range of the measurements for the species in the presently accepted sense, but this alone is not significant. The additional character given by Weber (1940: 413), viz. length-width proportion of the postpetiole ( faunulus and petiolatus with the postpetiole broader than long, " bigibbosus " with the postpetiole as long as broad), likewise does not seem to work. Moreover, petiolatus workers, as Weber (1938: 188) himself confesses, "are unsatisfactorily separated from the typical form" (= " bigibbosus "), the scape character given in his key (1940; 413) both contradicts the original description and a syntype specimen. The female of petiolatus , which unfortunately I did not see, is surely more distinct by its broader petiole and postpetiole. However, the evidence is not enough for recognizing a discrete form and 1 rather place petiolatus into synonymy of faunulus .

Variation. - Besides the differences in the female caste, as given for the race petiolatus , 1 have an even more striking variant from Manaus (DZSP n. 2198) which shows both in the worker and in the female caste a conspicuous pair of gibbosities on the anterior third of tergum 1 of gaster. This nest series agrees however with all other essential features of faunulus .

Bionomics. - Following is a brief digest of observations on faunulus bq Wheeler (1925: 45) and Weber (1946: 124-126), and on " bigibbosus " (= faunulus !) by Weber (1945: 14-16; 1946: 122-124), made in British Guiana and Trinidad.

C. faunulus is a rain-forest species. Its colonies are small, and the nests are usually found in rotted wood, but also in tangles of epiphyticroots, abandoned termite mounds and in the soil (clay stratum of an Atta mound!). The cavity size is variable, the volume varying approximately between 5-50 cc. The fungus garden is either sessile with lateral attachments to the wall, or pendant. The substrate consists of insect excrements, vegetable debris, seeds, woody fibers; it often includes parts of skeleton of ants (Ponerinae, Cephalotini), presumably used as framework. The bromatia are variable in aspect, their consistency transitional between those of Trachymyrmex and C. rimosus .

In captivity, C. faunulus specimens rejected dry chemical food such as hemoglobin, blood fibrin, egg albumen, dextrin, gelatin, peptose, maltose, diastase, but accepted bromatia from C. rimosus gardens grown on farine. Workers also lapped up eagerly the body juices of a crushed mosquito. They feign death when disturbed. Among inquilines have been registered a small milliped and an attophilous thysanuran.

Note. - The figures of the worker have been made from a very large worker taken by Mann at Covendo, Bolivia.