Okay, Kaye, I think I understand this report better thanks to you sending your sample chart off-list to me.
First, to explain the "gap"
I could see the "gap" in the line between two children who are in the same family on your chart. It is not a TMG issue. In fact, it is not really there . It is caused by the specific "zoom" value you are using to view your chart in Word. These lines are not "drawn" in this chart. Instead the document includes text characters which look like lines. Windows has to convert these individual line-drawing characters into pixels on the screen, and only does an approximate job. To cause the gap to "magically" disappear, change the zoom value. The Healy chart was being displayed at 132% zoom. Change that to 134% and you will see the gap disappear. Whether or not you see a gap when you print will also depend upon what zoom you use for printing.
Next, to explain illegitimate
You keep referring to "illegitimate" children, but their legitimacy really has no bearing on the issue. As I mentioned earlier, grouping of families is only based on the one or two parents linked as Primary to the child. In your own example of two children of the child "S", the first child "K" was born before the marriage of the parents (and thus was illegitimate) but the second child "R" was born after the marriage (and thus was legitimate). Yet TMG groups both children together equally within the single family of children born to that spouse. The legitimacy of a child has no bearing on the chart, only the child's linkage to its parents.
Finally, to explain the order of families
Thanks to your example chart I learned that my earlier description of sort order of families was inaccurate/incomplete. It turns out that if there is no Primary marriage tag with a date to use, TMG uses the earliest (sort) date of the Primary Birth group tag of any child in that family. Thus TMG works harder than I thought it did to try to get the list of separate families in an appropriate date order. The revised description based on my further testing is:
- All children are grouped together within the same family based on the spouse (or unknown spouse). All children who only have this one parent linked as their Primary parent will be grouped as a single "unknown spouse" family. (While they may actually be children of several different unknown spouses, TMG has no better alternative than to group them as one "family".) Each entire family group is sorted based on its own single "family" sort date.
- If there is a Primary marriage group tag for the parents, the (sort) date of that tag is used for the family sort date.
If there is a Primary marriage group tag, but it has no (sort) date, a blank date is used for the family sort date which will sort before non-blank family sort dates.
- If there is no Primary marriage group tag, the earliest (sort) date of a Primary tag in the Birth group of any child in that family is used as the family sort date. Children with no Birth group (sort) date will sort first within that family, but will not affect the sort order of the family unless no child has a non-blank Birth date.
If there is no Primary marriage group tag and no Primary Birth group (sort) date for any child in that family, a blank date is used for the family sort date which will sort before non-blank family sort dates.
If there are multiple families with the same date (or lack of date) for sorting, an order among those families has not been determined and should be considered random.
This should explain why your example child "S", who only has the mother linked and thus has an "unknown spouse", is listed before the subsequent family where there is a marriage group tag with a date and a spouse. This subsequent marriage group tag has a date, so that is used for its family sort date. The birth date of the child whose father is unknown is used for the sort date of the "unknown" family, which is a date earlier than the marriage tag date.
This is actually very clever of TMG, as it is more likely to put the various families in chronological order whether or not there is a marriage group tag with a date or even any marriages.
As best I can tell there is no problem with TMG here. There really is no gap in lines between children in the same family, and the order of families is the best TMG can do to get close to an appropriate chronological order. As a final note, I tested all of this in Version 6 through Version 9 with no differences.
Hope this helps explain,