Interaction in a Binomial how interpret

[Curious G.]

So, my doubt is simple, I ran this model:

glm(formula = pregnancy ~ Bull*SEMEN, family = binomial(),data = dt)

Just 2 bulls (a and b) and 2 semen (s and c) related to pregnancy. And I obtained these results:

      Deviance Residuals:

Min             1Q   Median       3Q      Max

-1.2377  -0.8748  -0.6172   1.1183   1.8720



Coefficients:

                        Estimate   Std.Error z value     Pr(>|z|) 
(Intercept)             0.1406     0.1878    0.749       0.454067 
Bulla                  -0.9038     0.2427   -3.724       0.000196 ***
SEMENs                 -1.7022     0.2749   -6.191       0.00000000596 ***
Bulla:SEMENs            0.9330     0.3798    2.456       0.014039 *

So, I made the exponential:

(Intercept)    Bulla          SEMENs         Bulla:SEMENs
1.1509434      0.4050290      0.1822767      2.5420404

For the Bullb has 60% more chance of pregnancy than Bullb, and SEMENc has 70% more chance than SEMENs. Ok, I know, I need consider just the interaction, but the result didn't make sense. The result should be negative too? Or is there a different way to interpret the interaction? Or this result didn't make sense in this analysis?
A table with additional information: