Vörös A. szerk.: Fragmenta Mineralogica Et Palaentologica 10. 1981. (Budapest, 1981)

0.86 (0.58-0.78), M

1 length 1.54-1.63 (1.36-1.60), M

2 length 1.59-1.68 (1.28-1.52), M

3 length 0.86-0.90 (0.72-0.88), M

1 width 1.95-2.23 (1.52-1.72), M

2 width 2.23-2.50 (1.78-2.04), M

3 width 1.86-2.08 (1.60-1.90), anteorbltal bridge 0.45-0.81 (0.42-0.70). Length of mandible 11.71-12.60 (10.15-12.30), C-M

3 6.71-7.40 (6.00-6.75), C-P4 2.64-2.90 (2.45-2.86), P4-M3 4.99-5.17 (4.404.90), M1-M3 4.06-4.54 (3.65-4.05), P

4 length 0.90-0.97 (0.76-0.90), P

4 width 0.68-0.77 (0.640.76), M

x length 1.36-1.50 (1.30-1.40), M

2 length 1.40-1.59 (1.30-1.48), M3 length 1.31-1.50 (1.20-1.36), M3 talonid width 0.72-0.81 (0.68-0.80), height of mandibular body under Mj 1.36-1.81 (1.31-1.59), height of mandibular body behind M

3 1.50-1.86 (1.31-1.50), height of coronoid process 3.28-3.82 (3.04-3.52) mm. Forearm length (estimated) 40.8-44.7 mm (MILLER

1 s M. nattererl 39-40.6, Hungarian specimens 37.8-41.6). Description and comparisons. Rostral portion of the skull In this new species is decidedly shortened and widened as compared to that of nattererl

. P

3 seldom In tooth-row line but pushed Inwards. Talon of P

4 strongly developed and with antero-internal cusp. Upper molars are especially stronger than in nattererl. Molars with no trace of protoconule as in emarginatus. As it is seen from the above measurements, its zygomatic width in the maximum range of our nattererl

. The same is true for interorbital constriction and the width between M

2-M . M. kretzolt sp.n. well exceeds nattererl in rostral width at outer crowns of M

3-M

3 and at C-C. There are some overlaps In upper C-M

3 and C-P

4 lengths, but not In P

4-M

3 and M*-M

3 lengths. Mandibular body of the new species behind M3 is especially stronger than that of nattererl. Ascending ramus of mandible with frontal margin perpendicular to coronoid process and not bent forward as In many nattererl

. In occlusal view angular process Is not bent buccally as in nat tererl, but keeps the main axis of the mandible. P3 frequently has a tendency to be pushed inwards from mandibular tooth-row line, or at least orocaudally pressed. There are some overlaps In lower C-M3 length of the new animal and that of M. nattererl. There Is no overlap, however, In the measurements of the lower molar-row lengths, just the minimum and the maximum values of the two ranges are the same. As regards the external measurements, fortunately a good number of fully preserved radii permit us to estimate the forearm length, which In many cases seems to be well over the forearm length of M. nattererl and equals that of M. bechstelnl Kuhl. The known measurements of M. kretzoll sp.n. suggest a generally bigger animal than M. nattererl

, with size of bechstelnl. but more robust than both of them. Its characters in some respects are more progressed than those of M. nattererl. Scatter diagrams (see, Figs. 1-4) made for this study clearly show that the rostral width measured at the outer edge of canines and especially at that of third molars are greater in the new species than those of M. nattererl. Regression lines of these characters are significantly different In M. nattererl and M. kretzoll. While the same Is true for C-P

4 length regarding their regression lines, the absolute measurements overlap here. There Is no significant difference between regression lines for M^-M

3 length (see, Fig. 3), although the absolute size difference is well shown. A discriminant analysis was carried out to test for any significant differences between the two species In a five-dlmensional morphological space (SVÁB 1979), (for the variables, see Figs. 1-4). Mahalanobls' distance was significant (D

2 = 3.6025, F = 3.53, p = 0.025). Individuals belonging to either of the two species are described by the discriminant function, Z • -1.49 xi + 2.84 x

2 + 1.33 X3 +2.18 X4 +2.04 X5. Frequency distributions of Z values of the two species are shown In Fig. 5. There is no overlap between the two species, but due to the small sample sizes, probability of misldentification calculated from the generalized distance (SVÁB 1979, p. 123) Is fairly high ( u = 0.95, p = 17.11). A brief analysis of the relative contributions of the various variables showed that mI-M

3 length and M

3-M

3 width are responsible for most of the difference between the two species (38.08% and 49.61%, respectively). Fig. 1. Scatter diagram correlating C-M

3 length with C-P

4 length in Myotis nattererl from Hungary (full circles) and Myotis kretzoll from Ördöglyuk Cave of Solymár (open circles), with regression lines and equations. Measurements in mm Fig. 2. Scatter diagram correlating C-M

3 length with M

3-M

3 rostral width in Myotis nattererl and Myotis kretzoll, with regression lines and equations. Legend, as for Fig. 1