Section I contains the names and purposes of all user-callable CML routines, arranged by GAMS category. Those unfamiliar with the GAMS scheme should consult the document "Guide to the SLATEC Common Mathematical Library". The current library has routines in the following GAMS major categories: A. Arithmetic, error analysis A3. Real A4. Complex A6. Change of representation C. Elementary and special functions C1. Integer-valued functions C2. Powers, roots, reciprocals C3. Polynomials C4. Elementary transcendental functions C5. Exponential and logarithmic integrals C7. Gamma C8. Error functions C9. Legendre functions C10. Bessel functions C11. Confluent hypergeometric functions C14. Elliptic integrals C19. Other special functions D. Linear Algebra D1. Elementary vector and matrix operations D2. Solution of systems of linear equations D3. Determinants D4. Eigenvalues, eigenvectors D5. QR decomposition, Gram-Schmidt orthogonalization D6. Singular value decomposition D7. Update matrix decompositions D9. Overdetermined or underdetermined systems of equations, singular systems, pseudo-inverses E. Interpolation E1. Univariate data (curve fitting) E3. Service routines F. Solution of nonlinear equations F1. Single equation F2. System of equations F3. Service routines G. Optimization G2. Constrained G4. Service routines H. Differentiation, integration H1. Numerical differentiation H2. Quadrature (numerical evaluation of definite integrals) I. Differential and integral equations I1. Ordinary differential equations I2. Partial differential equations J. Integral transforms J1. Fast Fourier transforms J4. Hilbert transforms K. Approximation K1. Least squares (L-2) approximation K6. Service routines L. Statistics, probability L5. Function evaluation L6. Pseudo-random number generation L7. Experimental design, including analysis of variance L8. Regression N. Data handling N1. Input, output N6. Sorting N8. Permuting R. Service routines R1. Machine-dependent constants R2. Error checking R3. Error handling Z. Other The library contains routines which operate on different types of data but which are otherwise equivalent. The names of equivalent routines are listed vertically before the purpose. Immediately after each name is a hyphen (-) and one of the alphabetic characters S, D, C, I, H, L, or A, where S indicates a single precision routine, D double precision, C complex, I integer, H character, L logical, and A is a pseudo-type given to routines that could not reasonably be converted to some other type.