Error in condition in Fortran

Question

Because the code:

    program acosseno
    implicit double precision(a-h,o-z)
PARAMETER(n=5)
    dimension x(n)
    dimension y(n)


alfa=0.6D0
k1 = 3
k2 = 5

print *, "Alfa is: ", alfa

open(1,file='a.dat') 


    x(1) = 5.0d0
x(2) = 10.0d0
x(3) = 2.0d0
x(4) = 20.0d0
x(5) = 30.0d0
y(1) = 1.0d0
y(2) = 3.0d0
y(3) = 15.0d0
y(4) = 8.0d0
y(5) = 35.0d0

do 1 i=1,n
    do 2 j = 1,n

if(i.ne.j)then
bi = sqrt(x(i)*x(i)+y(i)*y(i))
    bj = sqrt(x(j)*x(j)+y(j)*y(j))
ct = (x(i)*x(j)+y(i)*y(j))/(bi*bj)
endif

ang = acos(ct)  

    if(ang.le.alfa) goto 50 
    if(ang.gt.alfa) goto 60 

2       continue

50      print *, "le:", ang !ax = ang + k1 
60  print *, "gt:", ang !bx = ang + k2

    write(1,*)ax,bx

1       continue
    stop
        end

generates the following output:

 Alfa is:   0.59999999999999998     
 gt:   1.5707963267948966     
 le:   9.4061234627986545E-002
 gt:   9.4061234627986545E-002
 gt:   1.2408492346483417     
 le:  0.18311081726248352     
 gt:  0.18311081726248352     
 gt:  0.66477449481734552

Because for the same value (9.4061234627986545E-002) it is at the same time larger and equal to alpha? Other values are wrong too.