METHODS

Study Site
The study site is located on the right bank of the Rio Blanco, a tributary of the Rio Tahuayo, at approximately 4º40'S 73ºW (Fig. 1). It is characterized by highground primary rain forest ("tierra firme"). In addition to the two tamarin species, at least nine other primate species live in the area. The study site is provided with a grid of trails at intervals of 100 in, covering an area of about 1 kM2 . A detailed description of the site and its primate fauna is given by Bartecki and Heymann [1987a,b]

Subjects

This study focused on a mixed-species troop of three moustached tamarins, Saguinus mystax mystax (two adult males and one adult female), and six saddle-back tamarins, Saguinus fuscicollis nigrifrons (one adult male, one adult female, one young-adult female, one subadult male, one subadult female, and one infant/ juvenile male). All animals tolerated the approach of human observers to within 5 in. The moustached but not the saddle-back tamarins could be identified individually.

Data Collection

The study troop was observed for totally 417 h between July 1985 and May,1986 as part of an investigation of the function of long calls in moustached tamarins. While following the moustached tamarin group, the presence of the saddle-back tamarins was recorded at 5 min intervals. To be recorded as present, the saddle-back tamarins had to be within 20 in of the moustached tamarins.

Data on interspecific interactions were collected with an ad libitum sampling technique [Altmann, 1974]. Records included the type of interaction (agonistic, friendly), if possible the individuals involved, the context of the interaction, and the time of day. Agonistic interactions included avoidance (walking or running away on the approach of another individual) and aggression (threats, attacks, chases). Interactions were considered as having terminated when the animals involved in the interaction separated and/or took up another activity. Additional data on interspecific interactions were collected during 136 h between April 1986 and July 1986 during a study of the activity patterns and the scent-marking behavior of S. fuscicollis (Bartecki & Heymann, 1990). Interactions were not broken down into dyads; i.e., if a moustached tamarin simultaneously threatened two saddle-back tamarins, this was considered only one interaction. Therefore, the figures given are minimal estimates of the real number of interspecific interactions.

Long calls of moustached tamarins and saddle-back tamarins were recorded by noting the time of emission, location within the home-range, and if possible, the caller.

Information on mixed-species troops other than the study troop at the study site and its surroundings were collected opportunistically when the study group was not under observation.

Sleeping trees were marked and numbered. Their positions were determined to the nearest meter and to the nearest degree by using the trail system as reference points, and transferred to a map (scale 1:1,000). Distances between sleeping trees were read from this map. Time of leaving and retiring to a sleeping tree was determined to the nearest minute.

Data Analysis

For the analysis of time spent in association, the number of scans in which the species were associated was summed for hours (diurnal variation) or for complete days (comparison between days) and expressed as a percentage of the total number of scans of the respective time period. Variation of time spent in association was determined,by a two-way analysis of variance with the program BMDP 2V after arcsine transformation of percentages [Sokal & Rohlf, 1981]. For pair-wise comparisons of means a two-tailed t-test was applied, provided that a significant F ratio was found. Absolute frequencies were compared with the X2-test. To analyze whether long calling influenced the time that elapsed after leaving the sleeping tree until the two species joined each other in the morning, the time lag from leaving the sleeping tree until joining on mornings without long call was compared to the time lag from the first long call until joining on mornings when one or more long calls were given prior to joining. For this analysis only mornings on which the exact time of leaving the sleeping tree was known for moustached tamarins (n = 26 mornings) or saddle-back tamarins (n = 4 mornings) were included. Since these data were not normally distributed, comparisons were made with the Mann-Whitney U-test.

TABLE II.Sightings of Monospecific Groups and Mixed-species Troops of Tamarins in the Rio Blanco Study Area (Study Group Excluded)

^aMy S. mystax.
^bFu Saguinus fuscicollis .
^cMixed = mixed-species troop.