The gradient is . At the three specified points it is respectively
Each vector is perpendicular to its straight boundary edge and points out of the triangle, where increases from its negative interior values. The arrows in the preceding figure mark these directions.
The zero set consists of the three lines , and . Their pairwise intersections are
The bounded triangle has and . Its edges and interior contours are shown below.
Figure 1.
Negative contours inside the zero triangle, with outward boundary gradients and the interior minimum
. Details.
Set , so . The equation becomes . For its integrating factor is , giving
The initial value is , forcing . Hence
Substitution into the original equation verifies this answer throughout , including where the divided linear equation was singular. Its maximal connected interval through the initial point is .
Near the initial point, division by gives . An integrating factor is , and
Thus , and gives
Although division by was local, direct substitution verifies this smooth solution in the original equation for every real , including the zeros of .
On an interval avoiding zero put . A real solution requires , and the principal square root is . The homogeneous first-order differential equation reduces to
Where , separation of variables gives . Thus the nonconstant pieces are
restricted to intervals where the cosine of the phase is nonnegative. This sign condition is essential when substituting back into the original square root.
There are also the solutions and . They can join the nonconstant pieces with a continuous first derivative, because the transformed rate vanishes but is not Lipschitz continuous at . For example, on the positive axis every gives
The sticking solutions of a bounded square-root differential equation describe the same joining mechanism on either half-axis after using the increasing variable .
Multiplication by gives . Integrating and imposing yields . Therefore
The separable differential equation gives . Thus on , and the initial value gives . Hence
The initial-value branch cannot cross its singularity at zero.
For large , neighboring peaks overlap and the density has a broad, smooth, symmetric central hump, approximately on the scale . A precise central-profile statement follows from
for each fixed , almost surely and the integrand is bounded, so dominated convergence applies. The right panel of the preceding figure shows this Gaussian-looking core.
This approximation does not remove the heavy-tailed distribution. Here , so the symmetric distribution has no ordinary mean and no finite variance. Adding the independent Gaussian, which has finite absolute moments, leaves these conclusions unchanged. In particular a central Gaussian sketch cannot justify assigning variance to .
The density is the symmetric sum
For small , the normal distributions are narrow and well separated, giving peaks near each nonzero integer. The largest two peaks lie near and ; those near have approximately one quarter of their height, and the envelope falls as . There is no component centered at zero, and the density there is exponentially small. The left panel shows the resulting smooth comb.
Figure 1.
Narrow separated peaks and a broad central profile after Gaussian smoothing of heavy discrete tails
. Details.
Writing and comparing real parts and imaginary parts gives and . A real solution is impossible since . Thus , and positivity forces . Consequently and , giving
The right Argand diagram in the preceding figure shows representative points on the lines ; each row continues with spacing .
Multiplying by gives . The sixth roots of unity do not solve the original equation, since substituting either gives three. The other four do, because the extra factor is nonzero. Hence
Their real parts are and their imaginary parts are , giving the four corners in the left Argand diagram below.
Figure 1.
Four roots on the unit circle and the two horizontal rows of complex sine solutions
. Details.
Use the complex sine identity , then multiply by . The real part and imaginary part are
Multiply numerator and denominator by the complex conjugate of . For this gives
Thus its real part and imaginary part are respectively
The density is symmetric about , giving expected value . Its second moment is . Therefore
Equivalently, each equally weighted interval has variance and its centre is from the overall mean, giving by the law of total variance.
For , the five probabilities in increasing white-egg order are . The largest is , so the most common nest has three white eggs and one brown egg, .

There are unlisted articles, also show them or only show them.