Lisp_Float weirdness

struct Lisp_Float { struct lrecord_header lheader; union { double d; struct Lisp_Float *next; } data; }; #define float_next(f) ((f)->data.next) #define float_data(f) ((f)->data.d) alloc.c:: Lisp_Object make_float (double float_value) { [...] float_next (f) = ((struct Lisp_Float *) -1); float_data (f) = float_value; [...] } static Lisp_Object make_pure_float (double num) { [...] float_next (f) = ((struct Lisp_Float *) -1); float_data (f) = num; [...] } There are no other occurences of "float_next" in the code that I could find. Do you agree with me that the "next" field is useless ? If yes, I'll make the patch. OG.