1: \begin{abstract}

2: The gravitational radiation originating from a compact binary 

3: system in circular orbit is usually expressed as an infinite sum 

4: over radiative multipole moments. In a slow-motion approximation, 

5: each multipole moment is then expressed as a post-Newtonian 

6: expansion in powers of $v/c$, the ratio of the orbital velocity 

7: to the speed of light. The ``bare multipole truncation'' of the 

8: radiation consists in keeping only the leading-order (Newtonian) 

9: term in the post-Newtonian expansion of each moment, 

10: but summing over {\it all} the multipole moments. In the case of 

11: binary systems with small mass ratios, the bare multipole series 

12: was shown in a previous paper [Simone {\it et al.}~1997, 

13: {\it Class.~Quantum Grav.}~{\bf 14}, 237] to converge for all values 

14: $v/c < 2/e$, where $e$ is the base of natural logarithms. 

15: (These include all physically relevant values for circular inspiral.) 

16: In this paper, we extend the analysis to a ``dressed multipole

17: truncation'' of the radiation, in which the leading-order moments are 

18: corrected with terms of relative order $(v/c)^2$ (first post-Newtonian, 

19: or 1PN, terms) and $(v/c)^3$ (1.5PN terms). We find that the dressed 

20: multipole series converges also for all values $v/c < 2/e$, and

21: that it coincides (within 1 \%) with the numerically ``exact'' results 

22: for $v/c < 0.2$. Although the dressed multipole series converges, it is 

23: an unphysical approximation, and the issue of the convergence of the 

24: true post-Newtonian series remains uncertain. However, our analysis 

25: shows that an eventual failure of the true post-Newtonian series to 

26: converge cannot originate from summing over the Newtonian, 1PN, and 

27: 1.5PN part of all the multipole moments.

28: \end{abstract}

29: