"CgibbsSampler.MBnetwork.MBparams.MBgibbsPars" <-
function(x,params,gibbsPars,quiet=FALSE,sub=TRUE,prob=NULL,...)
  {
# Compatibility checks
    isCompat <- compatObjects(x,params)
    isValid <- validObject(gibbsPars)

# Variable inits
    if(gibbsPars@chains==1)
      margs.old <- matrix(nrow=(gibbsPars@chains+1),ncol=x@N)
    else
      margs.old <- matrix(nrow=gibbsPars@chains,ncol=x@N)
    states.old <- matrix(nrow=gibbsPars@chains,ncol=x@N)
    counts.old <- matrix(nrow=2*gibbsPars@chains,ncol=x@N)
    converged <- FALSE
    n <- 1

# Adjust 'dots' arguments for downstream functions;
# also, change verbosity depending on 'quiet' value
    RI.extras <- CS.extras <- list(...)
    RI.extras$verbose <- RI.extras$value <- NULL
    CS.extras$exclude <- CS.extras$prop <- NULL
    if(quiet) CS.extras$verbose <- FALSE

### BEGIN Convergence check

    while(!converged) {

# Burn-in
     if((n==1) || (gibbsPars@KLrepeat=="b")) {
      init <- TRUE
      if(!quiet) cat(paste("Starting burn-in, iteration ",n,":\n",sep=""))
      for(b in 1:length(gibbsPars@burn)) {
	nIter <- gibbsPars@burn[[b]]
	temp <- as.numeric(names(gibbsPars@burn)[b])
        for(c in 1:gibbsPars@chains) {
          if((n==1) && (b==1))
	    x <- do.call("randInit",c(list(x=x,sub=sub,prob=prob),RI.extras))
	  else {
	    x@States <- as.integer(states.old[c,])
	    x@Counts <- counts.old[c(2*c-1,2*c),]
	  }
	  if(!quiet) cat(paste("\t",nIter," iterations at temp ",temp," for chain ",c,"...",sep=""))
	  x <- do.call("CsampleNIter",c(list(x=x,params=params,n=nIter,method=gibbsPars@method,temp=temp,init=init,gsub=sub,prob=prob),CS.extras))
          if(!quiet) cat("Done\n")
          states.old[c,] <- x@States
	  counts.old[c(2*c-1,2*c),] <- x@Counts
        }
	if(init) init <- FALSE
      } # END Burn-in
      if(!quiet) cat(paste("Finished burn-in, iteration ",n,"\n",sep=""))
      if(gibbsPars@chains==1)
	margs.old[2,] <- x@Margs[2,]
     }

# Slicing
      init <- TRUE
      if(!quiet) cat(paste("Starting slicing, iteration ",n,":\n",sep=""))
      for(s in 1:length(gibbsPars@slice)) {
	nIter <- gibbsPars@slice[[s]]
	temp <- as.numeric(names(gibbsPars@slice)[s])
	for(c in 1:gibbsPars@chains) {
	  x@States <- as.integer(states.old[c,])
	  x@Counts <- counts.old[c(2*c-1,2*c),]
	  if(!quiet) cat(paste("\t",nIter," iterations at temp ",temp," for chain ",c,"...",sep=""))
	  x <- do.call("CsampleNIter",c(list(x=x,params=params,n=nIter,method=gibbsPars@method,temp=temp,init=init,gsub=sub,prob=prob),CS.extras))
	  if(!quiet) cat("Done\n")
	  margs.old[c,] <- x@Margs[2,]
	  states.old[c,] <- x@States
	  counts.old[c(2*c-1,2*c),] <- x@Counts
	}
	if(init) init <- FALSE
      } # END Slicing
      if(!quiet) cat(paste("Finished slicing, iteration ",n,"\n",sep=""))

# KL-divergence check
      kld <- kl.divergence(margs.old)
      if(!quiet) cat(paste("KL-divergence for ",gibbsPars@chains," chain(s) after iteration ",n," is ",kld,"\n",sep=""))
      n <- n+1
      if((gibbsPars@KLrepeat=="n") || (kld <= gibbsPars@thresh))
	converged <- TRUE
    }

### END Convergence check

# Sampling
    init <- TRUE
    if(!quiet) cat(paste("Starting sampling:\n",sep=""))
    for(s in 1:length(gibbsPars@samp)) {
      nIter <- gibbsPars@samp[[s]]
      temp <- as.numeric(names(gibbsPars@samp)[s])
      if(!quiet) cat(paste("\t",nIter," iterations at temp ",temp,"...",sep=""))
      x <- do.call("CsampleNIter",c(list(x=x,params=params,n=nIter,method=gibbsPars@method,temp=temp,init=init,gsub=sub,prob=prob),CS.extras))
      if(!quiet) cat("Done\n")
      if(init) init <- FALSE
    } # END Sampling
    if(!quiet) cat(paste("Finished sampling\n",sep=""))

    return(x)
  }