Dear Madam, Sir,
I would like to share some results on embryonic somatic muscle phenotypes of certain SMC holocomplex components in Drosophila melanogaster. These results will probably never see formal publication in a journal because my professor has recently retired and nobody will continue this line of research. 
The underlying experiments are documented in detail in my doctoral thesis (archiv.ub.uni-marburg.de/diss/z2018/0214/pdf/dmj.pdf, dx.doi.org/10.17192/z2018.0214).
Please note, I know some of my strains only by literature reference and am not entirely sure whether I matched them correctly with FlyBase IDs.
The embryonic musculature was visualized with an antibody stain against beta3-Tubulin, a muscle-specific protein that serves solely as a marker here (Leiss et al. 1988, FBrf0047887).
*--barrL305 and barrk14014 both show an identical embryonic somatic muscle phenotype: The muscle pattern is disturbed, with misshaped and disoriented or missing muscles and "holes" in the pattern; the pharynx musculature can appear distorted; heart cells are occasionally missing or duplicated. 
This phenotype varies in intensity both from embryo to embryo and from egglay to egglay: Some embryos have musculature that is close to the wild type in appearance, while other embryos have completely erratic muscle patterns and considerably reduced muscle mass and number of muscle nuclei.
*--The barren rescue construct UASP-Barren-eGFP III.2 (barrUASp.cOa.EGFP, FBal0284990; described in Oliveira et al. 2007, FBrf0201333) rescues this phenotype completely when expressed with 
*-a twist-Gal4 driver (from Baylies and Bate 1996, FBrf0086897; in FlyBase probably referenced as FBal0040490) 
*-or a DMef2-Gal4 driver (FBal0052385; described in Ranganayakulu et al. 1998, FBrf0104504).
The muscle phenotype is not rescued when UASP-Barren-eGFP III.2 is driven with
*-sns pro3-Gal4 (from Kocherlakota et al. 2008, FBrf0205177; probably FBal0230396)
*-or rP298-Gal4 (a kirre/dumbfounded-Gal4 driver from Menon and Chia 2001, FBrf0141462; probably FBal0305277).
*--Cap-G6 shows a phenotype similar to barren, which is however not variable but consistently strong. On top of the barren-like phenotype, the mutant embryos appear deformed, with incomplete dorsal closure.
*--Cap-G64 has a variable but often very strong phenotype, varying from a embryonic somatic muscle phenotype similar to Cap-G6 and the stronger barr phenotypes down to twisted embryos with no recognizable musculature. Intriguingly, transheterozygous Cap-G6/Cap-G64 embryos clearly show the Cap-G64 phenotype.
*--Cap-D2f03381 embryos have musculature like the wild type. A crossing of the deficiency lines Df(3R)BSC547 and Df(3R)BSC846, yielding transheterozygous embryos deficient for Cap-D2 and about 31 other genes in chromosomal area 99B5--99B10, shows no aberrations of embryonic somatic musculature or heart, either.
*--The Cohesin component mutants SMC1exc46 and vtd80Fh-1 show no embryonic somatic muscle phenotype.
I hope some of this information will be of use to someone.
With kind regards
Matthias Jacobs